Small arms casualties - tank crews

[Rokko]

8 minutes ago, IanL said:

Question for ya.  the number is the number of bullets it took to hit - correct?  What does the (k) mean?

One thing to consider is elevation.  Was the solider at the same elevation as the HT gunner?  The terrain for the solider test should be adjusted so that his height as closely matches the HT gunner's height.

No, restarting the game is *not* necessary.  If you save the setup with orders etc before the "Calculating" message is shown you can go back to that save as often as you like at watch a new version of the time line unfold.

What do you mean by correct? As I wrote in the procedural description I counted the delta in ammunition from the moment the hit happened in comparison to what was there at the beginning, 1008 x .30 rounds. If there was a round in midair (I could see the tracer) in the moment the round hit its target I discounted that round. So if the team had 987 rounds left when the target was hit and there was another round already fired that didn't cause the hit I'd count it as 20 rounds.

(K) stands for kill, everytime a "red" casualty occured I counted it as a kill. "Yellow" wounds (light ones) didn't occur in my 10 tests and would have been denoted by (W).

I didn't increase the elevation. The truck driver is kneeling behind the wall at the same elevation level as the halftrack. I guess you could put the wall on a small mound, but I don't think that would make difference. At best I'd expect this to decrease the accuracy, but probably it's not accounted for just as gun depression isnt' for tank guns.

[Battlefront.com]

7 hours ago, Rokko said:

Ok here's the test I have proposed in another thread on the same topic, now with a fleshed out design for peer review 

 

Excellent job laying things out.  Well done.

7 hours ago, Rokko said:

 

Edit: Following Steve's advice I haven't done more runs and posted my set up here first for discussion.

Smart, because it's got some serious problems.  The primary is the one you identified, specifically ducking behind the wall.  This creates huge problems when trying to do things like quantify rounds expended to get casualties.  In short, your test is not good for establishing if it's easier or harder to hit a HT gunner vs. a dude behind a wall.  Here's why:

The TacAI behavior is inherently different between the two targets.  The HT gunner tries to stay up so he can effectively use his weapon, the driver is predisposed to hiding.  So while you provided each one an effective place to hide from incoming fire, they are not coded to take advantage of it equally.  That pretty much crushes the test's validity right there because you are trying to get apples to apples results but you have an apple and a zucchini.  In other words, they aren't even both fruits

Often tests run into problems due to the artificiality of the conditions.  Remember, the game is coded on the assumption that nobody is trying to "rig the results" through direct manipulation.  That means when you change something to be unrealistic, you often destabilize the simulation itself.  This is a really tough thing to take into consideration, but it has to be.  In this case the HT gunner TacAI has a mission to protect the vehicle using the MG.  Because it has ammo and he has LOS he presumes that staying upright will allow him to fulfill his mission.  Therefore, he tries to remain upright because the TacAI expects that he can perform his job.  Which he can't because you're deliberately preventing him from doing so.  This artificially exposes the HT gunner to risk because the TacAI expects there's a potential reward for doing so, while at the same time the truck driver's TacAI has no delusions of grandeur and therefore drops at the first sign of trouble.

To put this into real world terms, giving two people similar math equations and timing how long it takes them to do them in their head runs into complications if one of the two people is a math wiz and the other struggles to come up with 5 when adding 2+3.  If the object of the test is to compare the speed of a math wiz vs. a math moron, that's fine.  But if the test presumes both are equal and therefore the results are directly comparable, then things run into problems very quickly.  This is akin to your test.

Conclusion, the test (as it is currently set up) is not capable of producing results which support your hypothesis.  In fact, the test is pretty much inherently predisposed to be in conflict with your hypothesis.

Steve

[Battlefront.com]

4 hours ago, Rokko said:

You're welcome. So far it seems the results could maybe (hope that's enough caveats ) indicate that indeed there is something with accuracy vs personell in the open and personell in vehicles. But maybe there is something off with the methodology or maybe there even is a reason why the accuracy is higher when shooting at halftrack gunners or TCs.

I wanted to comment on this separately.  Your question here is REALLY good.  That's the sort of thinking we want you guys to have when approaching tests like this.  CM is extremely complex and it is very difficult, even for us internal guys, to isolate factors correctly when examining certain behaviors.  Even the units selected can make a difference, for example a Panther G late is different than a Panther D.  Sometimes that difference doesn't matter, sometimes it's critical.  Depends entirely on what is being tested and how.

With that in mind, your suspicion that there might be different targeting methodology is correct.  The single truck driver is not perceived as a threat to the squad, whereas the HT gunner is.  This changes the squad's TacAI willingness to shoot.  Why?  Here's where we get into the difficulty of filtering out the simulation assumptions.  In this case one assumption is that firing your weapon draws attention to you and uses up valuable ammo, therefore firing is influenced based on perceived threat and/or payoff (a truck is low threat, but really high payoff).

Which means that your two squads are not inherently motivated to fire the same at the truck driver as they are the HT.  How much of a factor is it in this particular test?  Honestly I don't know, but I suspect not much because the previous problem I outlined most likely makes this particular one irrelevant because the driver is destined to use the cover he's been given.  If you removed the wall, then this issue might become more relevant because the opportunity to keep shooting at the driver remains open.

Steve

[Battlefront.com]

6 hours ago, JoMc67 said:

In my experience in CMx2, it's about 1 sec ( enough time for the Crew to take their last breath ).

Comments like this aren't helpful because this is a "feeling" piling onto a "feeling".  And since I know what you just said is factually incorrect, I also know that we don't need to run tests to know that

6 hours ago, JoMc67 said:

That's just it, how long should it take to be spotted at that range...Me thinks a Spotting Cycle or three...But, Steve, thinks at 200 meters the Crew should be "Slaughtered" in mere seconds (I think more like 25-40 meters), and so it doesn't matter really...

And more misscharacterizing and hyperbole doesn't help either.  Honestly, sometimes I don't know why I bother spending my precious time on this Earth trying to have a productive and rational discussion with you guys, in an honest attempt to improve the game, when I get painted as being a pig headed dunce who just wants to move onto something else.  If that were the case I'd just say "you're all wrong", lock the thread, and move on... wouldn't I?  If that's the sort of response you would prefer from us, keep it up and I'll try to conform to your perception so you have something to compare against.

The context of my comments seems to have slipped your mind or gone right of your head.  Therefore I will remind you what I actually said:

As for Chris' test it had OPTIMAL conditions for the infantry, so they spotted very quickly at 200m.  The optimal conditions?  Bright sun, flat ground, nothing between the tank and the Squad, nothing else for the Squad to get distracted by, tank not shooting at all, etc.  In other words, a situation that should not be frequently seen in CM.

As stated in my previous two posts, when you start artificially messing with a sim you should expect the results will be different than when the sim is allowed to function as intended.  Stripping away all distractions from the Squad should result in near instantaneous spotting.  In real life it would as well, would it not?  Or do you think if you put 11 soldiers on a firing line they wouldn't instantly spot the rather small piece of paper with the black circle in the middle?  I mean seriously, do you really expect a trained soldier to sit around for 7 seconds thinking "gee, I wonder what I'm supposed to shoot at?".  If that were the case then I'd say some nation needs to sack ever officer and NCO connected to infantry training and start from scratch

Which is to say that when people deliberately make an artificial test scenario they have to be prepared to NOT read much into it other than the very specific thing being examined.  If you want to see how long it takes to spot a TC by an average squad in different combat circumstances, then you need to design a test specifically for that.  As you can see from my critique of Rokko's test, it's hard enough to get test conditions that match what you match what is being tested.

Steve

[JoMac]

Actually, I was about to respond to 'Rokko', but Steve responded with the same and in more detail.

Now, think we might have an idea of why there is that 'Bullet Magnet' theory towards Vehicles as opposed to other units.

Ok, Steve, you win...I'm just glad you, me and others are at least having these types of discussions....and so, I have nothing further to add 

Edited April 11, 2016 by JoMc67

[A Canadian Cat]

39 minutes ago, Battlefront.com said:

The single truck driver is not perceived as a threat to the squad, whereas the HT gunner is.  This changes the squad's TacAI willingness to shoot.  Why?  Here's where we get into the difficulty of filtering out the simulation assumptions.  In this case one assumption is that firing your weapon draws attention to you and uses up valuable ammo, therefore firing is influenced based on perceived threat and/or payoff (a truck is low threat, but really high payoff).

I was thinking that @Rokko's method of counting bullets to get a kill vs time might be a good measure.  Mind you that would also mean you really want one shooter to make it fare because after all if one or two guys take pot shots at the guy behind the wall while every body tried to light up the HT gunner even counting bullets is not going to work.

Perhaps the shooter should be a one person vehicle crew member as well.

[A Canadian Cat]

1 hour ago, Rokko said:

What do you mean by correct? As I wrote in the procedural description I counted the delta in ammunition from the moment the hit happened in comparison to what was there at the beginning, 1008 x .30 rounds.

I was asking for verification.  For some reason when I wrote that in my head it was not clear.  After re reading your post I have no idea why it was unclear in my head - it just was.

[A Canadian Cat]

56 minutes ago, Battlefront.com said:

The TacAI behavior is inherently different between the two targets.  The HT gunner tries to stay up so he can effectively use his weapon, the driver is predisposed to hiding.  So while you provided each one an effective place to hide from incoming fire, they are not coded to take advantage of it equally.

What if the driver behind the wall was given a target order some where away from the test shooters.  Would that give him the same motivation to stay up and do his job as the HT gunner? My guess is no but I am curious.

[Battlefront.com]

56 minutes ago, JoMc67 said:

Actually, I was about to respond to 'Rokko', but Steve responded with the same and in more detail.

Now, think we might have an idea of why there is that 'Bullet Magnet' theory towards Vehicles as opposed to other units.

Radiomen, flamethrowers, and other things also rank higher in the pecking order in terms of motivation to target.  But other factors such as range, LOF strength, exposure, etc. all come into play as well.  There's so much going on under the hood that it's very difficult to isolate things in a way that produces a meaningful result.  At least sometimes, other times it's fairly straight forward.  Even something seemingly simple like "how much cover does a wall offer" has to be carefully considered because there is no such thing as a singular wall concept.  Is the shooter higher than the target behind the wall?  Is the wall on sloped ground or flat?  Are there obstacles on the defending side of the wall that prevent soldiers from easily taking cover?  What happens if soldiers are between a building and a wall, does that change the dynamics?

This is why we're still having discussions like this after nearly 10 years of a playable CMx2 game engine.

(oh, and to answer a previous missive... there has been NO changes to any of the core game mechanics in terms of spotting, targeting, etc. between CMFB and the last batch of CM games.  In fact, CMBN v3.12 uses the same exact code)

56 minutes ago, JoMc67 said:

Ok, Steve, you win...I'm just glad you, me and others are at least having these types of discussions....and so, I have nothing further to add 

Yup, having these discussions shows why there's reason to have these discussion   If we were playing a traditional 2D hex game we certainly wouldn't.

32 minutes ago, IanL said:

I was thinking that @Rokko's method of counting bullets to get a kill vs time might be a good measure.  Mind you that would also mean you really want one shooter to make it fare because after all if one or two guys take pot shots at the guy behind the wall while every body tried to light up the HT gunner even counting bullets is not going to work.

Perhaps the shooter should be a one person vehicle crew member as well.

Counting bullets only has value if the counts represent something meaningful.  With the test set up the way it is, that doesn't seem to be the case.

28 minutes ago, IanL said:

What if the driver behind the wall was given a target order some where away from the test shooters.  Would that give him the same motivation to stay up and do his job as the HT gunner? My guess is no but I am curious.

Correct, the answer is "no".  The reason why is inherently that dude wants to duck.  Period.  You can make him Crack Experience and Fanatic modifiers and it doesn't matter... ducking is in his DNA in a way it is not in the DNA of the HT gunner.

I also was thinking "what would I substitute for that truck driver" and I thought of another HT but this time facing backwards.  This radically alters the gunner's sense of threat to self and that should produce a very different result.  It would at least show how much the shield is entering in the equation.  Then I realized this is slightly flawed as well because any shot to an unprotected rear of a soldier puts the dude into instant Red Alert mode and therefore likely have similar behavior as the truck driver EXCEPT he'd be more likely to pop back up (I think).  Then I realized that there could be a problem with the vehicle's backend giving the gunner a bit more protection than from the front if using a 251.  A 250 might be more equal front to back.  In thinking about it more, shooting at the side of a HT might be slightly better than the rear.

Even with that suggestion on the table I'm not sure it's the right way to go.  I'd have to think about it more.

Steve

[sttp]

But the tests are measuring shooting accuracy with respect to number of bullets expended, not time 'til KIA / WIA, right?

It may not be as perfect a test as we would all like, but it's hard to imagine how one could simplify variables more than Rokko has done here. I'd maybe add a few different ranges, and try it with different vehicle types and soldier types before drawing too many conclusions, but clearly, something strange is going on.

For my 5 trials, for whatever they're worth, every HT gunner was a KIA after 11.8 shots, on average. Every foot soldier was a WIA after 67 shots, on average. The idea, of course, is to combine the numbers from all players' trials and then do the stats, but I thought I'd post these for, again, whatever they're worth.

My "raw" results with Rokko's test:

vs. German halftrack
1 - 5(K)
2 - 6(K)
3 - 12(K)
4 - 24(K)
5 - 12(K)

vs. German soldier
1 - 104(W)
2 - 62(W)
3 - 33(W)
4 - 107(W)
5 - 29(W)

If this particular test isn't one that can reveal the difference everyone seems to be talking about, then why don't we all, together, construct the kind of tests that would be most useful. Even if it required 30 trials each for testing, say, 5 or 6 variables.

 

 

Edited April 11, 2016 by sttp

[sttp]

How about 3 different infantry types vs. 3 vehicles types, then each of those same 3 infantry types vs. 3 other soft target types? At 1 or 2 different distances? That would be very useful, but... well over 1000 trials total, Twice that if we do 2 different distances, But if we break it up into 20 people... maybe? It'd be tedious, sure, but very doable. And this issue, after dozens of long threads arguing both sides, could be put to rest.

If the 3 by 3 types is too daunting, then how about 2 types by 2 types, at 2 different distances. (Thay'd be 240 total trials, I think.)

[Battlefront.com]

1 hour ago, sttp said:

But the tests are measuring shooting accuracy with respect to number of bullets expended, not time 'til KIA / WIA, right?

Correct, but the fact that the soldier is predisposed to hiding means that inherently it will take more bullets to cause a casualty because for the most part he's prone behind a 100% bullet shield, but the squad still shoots at it for suppression (correct?).  This means you can produce an interesting examination of one or the other, but you can not compare the two against each other in any meaningful way.

1 hour ago, sttp said:

It may not be as perfect a test as we would all like, but it's hard to imagine how one could simplify variables more than Rokko has done here. I'd maybe add a few different ranges, and try it with different vehicle types and soldier types before drawing too many conclusions, but clearly, something strange is going on.

And this is why it's important to have vetted testing conditions match what you're trying to test for.  If I counted how many shots from a 57mm ATG it takes to kill things and found it takes 5 shots on average to kill a King Tiger and 1 shot to kill a truck in order to compare which is easier to kill, would you conclude "there is something strange going on"?  I sure hope not   The test that Rokko set up, and you continued with, is about the same. 

1 hour ago, sttp said:

If this particular test isn't one that can reveal the difference everyone seems to be talking about, then why don't we all, together, construct the kind of tests that would be most useful. Even if it required 30 trials each for testing, say, 5 or 6 variables.

As I said, this is not so easily done.  I did make some suggestions that should at least make the two situations more comparable.  Specifically, a SPW 250 facing forward and a SPW 250 facing 90 degrees to the shooter.  This gives some sense of how much protection the shield is offering.

However, I'm still trying to figure out what exactly people are seeing that they aren't expecting to see?  I think we've already identified a few possible tweaks to realistically increase TC survival when under fire.  Other than that I really don't know what people think is "wrong" because logically 200m or less should be suicide for any TC that doesn't button up very quickly (hence the thing we're looking at tweaking is speed of buttoning) and should stay buttoned (the other thing we're looking at).

Steve

[Rokko]

34 minutes ago, Battlefront.com said:

Correct, but the fact that the soldier is predisposed to hiding means that inherently it will take more bullets to cause a casualty because for the most part he's prone behind a 100% bullet shield, but the squad still shoots at it for suppression (correct?).  This means you can produce an interesting examination of one or the other, but you can not compare the two against each other in any meaningful way.

And this is why it's important to have vetted testing conditions match what you're trying to test for.  If I counted how many shots from a 57mm ATG it takes to kill things and found it takes 5 shots on average to kill a King Tiger and 1 shot to kill a truck in order to compare which is easier to kill, would you conclude "there is something strange going on"?  I sure hope not   The test that Rokko set up, and you continued with, is about the same. 

As I said, this is not so easily done.  I did make some suggestions that should at least make the two situations more comparable.  Specifically, a SPW 250 facing forward and a SPW 250 facing 90 degrees to the shooter.  This gives some sense of how much protection the shield is offering.

However, I'm still trying to figure out what exactly people are seeing that they aren't expecting to see?  I think we've already identified a few possible tweaks to realistically increase TC survival when under fire.  Other than that I really don't know what people think is "wrong" because logically 200m or less should be suicide for any TC that doesn't button up very quickly (hence the thing we're looking at tweaking is speed of buttoning) and should stay buttoned (the other thing we're looking at).

Steve

My observation in the test file was that the truck driver would only duck away for very short periods of time and while they were "gone" the firing ceased. So I am not sure that the test is inherently invalid. Of course the cowering in some way set back the aiming progress so to say but not completely. All shots were essentially aimed shots. Since the trucker guy was an elite fanatic maniac he'd usually get right up again.

[Sgt Joch]

were TC's bullet magnets?, you can find anecdotes to support any viewpoint:

"To protect CCA's open right flank, Gaffey ordered Col. Wendell Blanchard to form the Reserve Combat Command as a balanced task force (using the 53d Armored Infantry Battalion and 37th Tank Battalion) and advance toward Bigonville. Early on 23 December CCR left Quatre-Vents, followed the main road nearly to Martelange, then turned right onto a secondary road which angled northeast. This road was "sheer ice" and much time was consumed moving the column forward.

About noon the advance guard came under fire from a small plot of woods near a crossroads at which point CCR would have to turn due north. The accompanying artillery battalion went into action, pouring high explosive into the woods for nearly an hour. One rifle company then dismounted and went in to clean out the survivors. The company found no serious resistance, returned to the road, and was just mounting its half-tracks when a fusillade of bullets burst from the little wood. Apparently the enemy had withdrawn during the shelling, only to return at the heels of the departing Americans. Tanks were now sent toward the crossroad but were stopped by mines. All this had been time-consuming. Bigonville was still a mile away, and Blanchard ordered a halt. The enemy in the woods continued to inflict casualties on the troops halted beside the road. Even the tankers were not immune-nearly all of the tank commanders of one company were picked off by rifle fire.

In the course of the night the Germans left the wood and fell back to the shelter"

Cole, "Ardennes", p. 530

http://www.history.army.mil/books/wwii/7-8/7-8_21.htm#p530

Edited April 11, 2016 by Sgt Joch

[Battlefront.com]

10 minutes ago, Rokko said:

My observation in the test file was that the truck driver would only duck away for very short periods of time and while they were "gone" the firing ceased.

OK, I wasn't sure if there would be stray shots.

10 minutes ago, Rokko said:

So I am not sure that the test is inherently invalid. Of course the cowering in some way set back the aiming progress so to say but not completely. All shots were essentially aimed shots. Since the trucker guy was an elite fanatic maniac he'd usually get right up again.

It does matter.  There is an inherent difference between repeated ducking and popping back up and staying up.  It's true in real life and it's true in the game.  Which is why we're looking into trying to move some of that behavior into the vehicle crew TacAI, because that would be a good thing to do.

Anyway, my point remains as it has been explained already.  You can compare apples to oranges only if you are taking their respective differences into account.  I do not believe your test does and therefore, as it is, doesn't produce results which offer much in terms of analysis.

6 minutes ago, Sgt Joch said:

were TC's bullet magnets, you can find anecdotes to support any viewpoint:

"To protect CCA's open right flank, Gaffey ordered Col. Wendell Blanchard to form the Reserve Combat Command as a balanced task force (using the 53d Armored Infantry Battalion and 37th Tank Battalion) and advance toward Bigonville. Early on 23 December CCR left Quatre-Vents, followed the main road nearly to Martelange, then turned right onto a secondary road which angled northeast. This road was "sheer ice" and much time was consumed moving the column forward.

About noon the advance guard came under fire from a small plot of woods near a crossroads at which point CCR would have to turn due north. The accompanying artillery battalion went into action, pouring high explosive into the woods for nearly an hour. One rifle company then dismounted and went in to clean out the survivors. The company found no serious resistance, returned to the road, and was just mounting its half-tracks when a fusillade of bullets burst from the little wood. Apparently the enemy had withdrawn during the shelling, only to return at the heels of the departing Americans. Tanks were now sent toward the crossroad but were stopped by mines. All this had been time-consuming. Bigonville was still a mile away, and Blanchard ordered a halt. The enemy in the woods continued to inflict casualties on the troops halted beside the road. Even the tankers were not immune-nearly all of the tank commanders of one company were picked off by rifle fire.

In the course of the night the Germans left the wood and fell back to the shelter"

http://www.history.army.mil/books/wwii/7-8/7-8_21.htm#p523

Yup.  One of the engagements I remember, unfortunately not in detail, was in the early stages of the US Army moving on Aachen.  Along the advance route the Germans put out a bunch of rifleman and they knocked off so many TCs that the attack had to be cancelled and the armor withdrew.

Steve

[Rokko]

4 hours ago, Battlefront.com said:

 

I'd like to reply in more detail because I still don't believe it is an inherently invalid comparison I am trying to make. But I dont have access to a PC or laptop atm so this has to suffice for now.

I dont know if you have checked out the test files yourself, but in each file there only one pair of shooters and one target each. So things like pecking order or target priority should not matter at all. Also the shooters are firing on their on, it's a hands off test. I haven't specifically measured ROF but it didn't seem to differ for each target.

I absolutely agree that the ducking away matters, since it hinders effective zeroing in. Unfortunately there is nothing I can do about, I'd have to be able to completely turn off the Tac-AI. I'd again suggest you open up the file and see for yourself, the target the HT gunner presents is tiny from the POV of the shooters, it is essentially only the top of the helmet. In comparison the truck guy presents a huge target.

Maybe I could do a little hack and replace the cowering animation to be the same as the kneeling animation, but that would be a shot into the blue since I don't really know what the cowering status itself might do beside the animation switch.

Also, to me this whole thing is not at all about whether the gun shield works as intended, it's about finding out if there is some technical issue that makes the accuracy calculations more, well accurate when applied to shooting at personell in vehicles compared to shooting at infantry in the open.

If that were the case there'd be two options I think, either accuracy vs troops in the open is too low or accuracy against vehicle people is too high. In any case there'd be a difference I have trouble seeing a justification for. Why should a soldier shoot better at a meat bag standing behind a shield in a vehicle than he does at a larger (percieved) meat bag behind a stone wall.

Also I don't see how the comparison of an ATG shooting at a Tiger vs shooting at a truck holds true since I am comparing a meat bag and a meat bag. Only that one meat bag keeps ducking away for a few seconds at a time.

So would you consider the test as valid if it were possible to deactivate the self-preservation?

[Rokko]

5 hours ago, IanL said:

I was thinking that @Rokko's method of counting bullets to get a kill vs time might be a good measure.  Mind you that would also mean you really want one shooter to make it fare because after all if one or two guys take pot shots at the guy behind the wall while every body tried to light up the HT gunner even counting bullets is not going to work.

Perhaps the shooter should be a one person vehicle crew member as well.

I've actually tried this before, but truck drivers tend to carry too little ammo for these test, they too often run out off ammo before hitting anything while MG ammo bearers have more than enough.

I'll check if is possible to reduce their headcounts. Overall I dont think it matters though, as long as you keep the number of shooters the same for each test.

[Lethaface]

No ammo in the truck?

[Michael Emrys]

On April 10, 2016 at 3:08 AM, niall78 said:

I used to take an awful lot of TC casualties. I don't these days. The trick? I try never to expose them to any kind of effective small arm fire. That means never exposing them to any small arms fire under 500 metres.

Same here. Back in the early days of BN I noticed that I was losing more TCs than I liked, so I started buttoning them up as soon as I thought they were within small arms range of likely enemy positions, which in the majority of the games I play meant from the start of the game. I think it may be years since I lost more than one or two TCs.

Michael

[Michael Emrys]

On April 10, 2016 at 4:49 AM, niall78 said:

I wonder does the use of half tracks and TCs differ between people that play campaigns heavily and those that play one-off games heavily?

An interesting question for sure. But in my own case the answer would be no. Force preservation is always a serious goal for me. If it is possible, I would greatly prefer to win without losing a single man. That appears to be at least extremely unlikely in CM, but I try to approach that goal as closely as I can. If I ever have more than 5% casualties, I feel like I screwed up somewhere, and with benefit of hindsight I usually know exactly where.

As an example, the first time I played the tutorial, I lost an AC and a Sherman and a bunch of men shot up. I was so disgusted I quit halfway through and started over. That I had learned from my mistakes was shown by the fact that in this second playing I won a total victory losing only 3 men KIA and 3 WIA.

Michael

[bwgulley]

I don't have an issue with TC's being shot. I do have a problem with Stug TC's opening up well with in small arms fire range and getting shot. My last game I had the 105 Stug at 250m with direct LOS to infantry open up and fire the mg. Of course he was shot, crew panicked and jumped out. Why in God's name would you expose your self to fire when you have a 105m gun with clear LOS? Perhaps I'm wrong but isn't the point of an assault gun to roll up to darn near point blank and blast away with heavy artillery. Frankly I don't play RT anymore because of this issue. I see the same happening here. Damn shame. Customer since CMBO pre order.

[niall78]

1 hour ago, bwgulley said:

Perhaps I'm wrong but isn't the point of an assault gun to roll up to darn near point blank and blast away with heavy artillery.

It might be if you were tackling a heavily suppressed bunker or house. Any other time I would suggest such a tactic is a death sentence for the assault gun and its crew.

Putting any kind of a tank, assault gun or half-track into close contact with unsuppressed infantry is asking for losses for little result.

[Battlefront.com]

3 hours ago, bwgulley said:

I don't have an issue with TC's being shot. I do have a problem with Stug TC's opening up well with in small arms fire range and getting shot. My last game I had the 105 Stug at 250m with direct LOS to infantry open up and fire the mg. Of course he was shot, crew panicked and jumped out. Why in God's name would you expose your self to fire when you have a 105m gun with clear LOS? Perhaps I'm wrong but isn't the point of an assault gun to roll up to darn near point blank and blast away with heavy artillery. Frankly I don't play RT anymore because of this issue. I see the same happening here. Damn shame. Customer since CMBO pre order.

Every game since CMBO has had problems, sometimes major problems.  This is a rather small and limited one which we are looking into fixing.  The problem has always been there, however something about the CMFB setting seems to be bringing out the worst in it.  I've speculated it's because StuGs are more common and the terrain is tighter.

Having said that, this isn't 100% a game behavior issue...

1 hour ago, niall78 said:

It might be if you were tackling a heavily suppressed bunker or house. Any other time I would suggest such a tactic is a death sentence for the assault gun and its crew.

Putting any kind of a tank, assault gun or half-track into close contact with unsuppressed infantry is asking for losses for little result.

Exactly.  The primary cause of this is players moving armor into close range without proper supporting arms and suppressive fire.  Stay further back and even with the current TC unbutton behavior things should be fine.  The close proximity to infantry causes the TC to be *more* likely to man the MG (current logic) and at the same time the enemy infantry is extremely dangerous.

Since point blank range engagements like this were to be avoided in real war, so too should the player.

Steve

[Battlefront.com]

5 hours ago, Rokko said:

I'd like to reply in more detail because I still don't believe it is an inherently invalid comparison I am trying to make. But I dont have access to a PC or laptop atm so this has to suffice for now.

I dont know if you have checked out the test files yourself, but in each file there only one pair of shooters and one target each. So things like pecking order or target priority should not matter at all. Also the shooters are firing on their on, it's a hands off test. I haven't specifically measured ROF but it didn't seem to differ for each target.

Targets come with different "incentives" for enemies to shoot at it. By "pecking order" I meant that infantry is more incentivized to fire as fast and as much as possible against something like a flamethrower than a truck driver.  How much of a difference it makes in this artificial setting... I have no idea, but it's something to take into consideration.  You can get high priority target such as a single flamethrower dude into the game by using the pre-battle casualties to bump off the assistant.

Quote

I absolutely agree that the ducking away matters, since it hinders effective zeroing in. Unfortunately there is nothing I can do about, I'd have to be able to completely turn off the Tac-AI. I'd again suggest you open up the file and see for yourself, the target the HT gunner presents is tiny from the POV of the shooters, it is essentially only the top of the helmet. In comparison the truck guy presents a huge target.

We've had discussions about accuracy many times before.  Sometimes it's better to be out in the open as it makes aiming for a specific spot harder.  Hitting a specific spot, on the other hand, is a separate issue.  And this is where the dude bouncing up and down has a big impact.  When the truck driver ducks down the riflemen remove their rifle from the aim point, which means each time the truck driver comes back up they have to target fresh.  At best they might be adjusted for range and shooting conditions compared to the unbroken engagement against the HT driver which allows for better accuracy of follow up shots.  Now, in a real world situation the difference might not be all that huge because soldiers targeting the truck driver might keep their weapons trained on that position.  CM, however, does simulate this because 99% of the time it's not what happens in combat.  Target goes down, soldiers start to look for other things to target because the world is a hostile place and target fixation, especially if you think you got the guy, isn't a healthy practice.

Quote

Maybe I could do a little hack and replace the cowering animation to be the same as the kneeling animation, but that would be a shot into the blue since I don't really know what the cowering status itself might do beside the animation switch.

Cowering is a "state" and therefore changing the animation won't change the behavior.

Quote

Also, to me this whole thing is not at all about whether the gun shield works as intended, it's about finding out if there is some technical issue that makes the accuracy calculations more, well accurate when applied to shooting at personell in vehicles compared to shooting at infantry in the open.

In real life I would expect the chances of placing a shot onto the shield of the HT would be easier than getting a shot similarly placed relative to a guy kneeling behind a wall.  At least in terms of combat type shooting.  Whether it's more effective, however, is a second question.

Quote

If that were the case there'd be two options I think, either accuracy vs troops in the open is too low or accuracy against vehicle people is too high. In any case there'd be a difference I have trouble seeing a justification for. Why should a soldier shoot better at a meat bag standing behind a shield in a vehicle than he does at a larger (percieved) meat bag behind a stone wall.

This is why you should swap out the Squads with Snipers.  If you're looking at answering this question you need to be looking at one shot one effect.  It also likely reduces the chances the driver will duck down because one bullet zinging by is much less of a threat than 11 bullets zinging by.

Quote

Also I don't see how the comparison of an ATG shooting at a Tiger vs shooting at a truck holds true since I am comparing a meat bag and a meat bag. Only that one meat bag keeps ducking away for a few seconds at a time.

That's what I've been trying to get you to realize... you are not comparing apples to apples.  You are comparing one guy who is predisposed to standing up and taking on enemy fire (because he assumes he can fire back) to another guy who is instructed to take cover.  This alters the chances of hitting and that skews the data.  You can not ignore this any more than you can ignore factors such as the angle a target tank is facing in one situation compared to another.

Put another way, because we know these factors DO have an impact on the results, you can't use comparative results without minimizing the differences.  It is just not possible to do, therefore look for another way that doesn't run into this problem. 

Quote

So would you consider the test as valid if it were possible to deactivate the self-preservation?

It would be pretty close, probably yes.  But we can not turn off the TacAI and therefore you have to find another way to do a comparison.  Arguing with me on this point won't further this discussion, so move onto something that will.

To summarize my suggestions for a new test, I'd do the following:

Make 12 "lanes" with Tall Walls separating each.  For this test I'd go with something like 100m wide and each lane a minimum of 600m deep.  Which means you could do a map that is 1200x600 and get what you need.  Run a strip of different dirt every 100m from "left to right" to mark the distances.

Lanes 1-3 and 7-9 put a SPW 250 with Regular or Veteran Crew and neutral modifies.  For the first group put them at 200m and for the second group put them at 300m

Lanes 4-6 and 10-12 put depleted flamthrowers (i.e. 1 man) at the 200m and 300m marks respectively.  Place a short wall between them and the firing line.

On the firing line for each late place a depleted sharpshooter (i.e. 1 man) with Regular or Veteran experience and neutral modifiers. 

Give each Target a short Cover Arc, give each Shooter no orders at all.  Hit GO! and note the following information for each lane:

1.  Shots fired before Target becomes casualty or ducks down or becomes a casualty.  Note which is the case.

2.  If a Target pops back up, note it and repeat the same tabulation as if it is a brand new example.

3.  Continue until all Targets are eliminated.

4.  Tabulate how many shots were needed to affect a state change for the Target and how many rounds were necessary to make it happen.  Group these results by state change (Duck, WIA, KIA).  Treat

Note that in my methodology I am not tabulating the data to have a total shot count for Targets that bounce up and down.  Instead I am treating each as a separate incident.  With the data recorded that way I can examine it individually or combine it for Target totals.  Which is a reminder that HOW you record the data is as important as the data itself.

As for the Experience settings, I chose neutral "game average" so the results are more likely to be comparable to common game situations.  Making them Elite and Fanatical means you're testing for something that isn't seen on a regular basis and is therefore not as comparable.

I can't think of anything significantly wrong with this approach, so I think it's a pretty solid way to go.

Steve

[Heirloom_Tomato]

12 hours ago, Battlefront.com said:

With that in mind, your suspicion that there might be different targeting methodology is correct.  The single truck driver is not perceived as a threat to the squad, whereas the HT gunner is.  This changes the squad's TacAI willingness to shoot.  Why?  Here's where we get into the difficulty of filtering out the simulation assumptions.  In this case one assumption is that firing your weapon draws attention to you and uses up valuable ammo, therefore firing is influenced based on perceived threat and/or payoff (a truck is low threat, but really high payoff).

 

In another thread on halftrack gunners dropping like flies, I ran a few tests and came to the conclusion that the game is having infantry not only spot for enemy targets but also do a threat analysis of the targets they spot. I hypothesised that a halftrack gunner or TC is a higher threat than an infantryman in the open due to the amount of "hurt" they can send in the direction of my men. Therefore, my infantry are more likely to all fire at once at a halftrack gunner or TC out of concern for their own safety, rather than to shoot at an easier to hit target. Is your quote above confirming what I suspected?

If so, it would seem to me, that players tactics are causing higher death rates rather than faulty game mechanics. With regard to the OP complaint about additional crew members unbuttoning and being killed off, a tweak to the system to reduce their willingness to do so would be great. I know I would personally be not very keen to stick my head out of a hatch that I just saw my crew mate get wounded or killed in.