German tank vulnerability

[Vergeltungswaffe]

Question, if a tank placed at an angle to it's opponent and not just straight inline at him, in reality it should create higher chances of bouncing off the armor. Does this effect apply to our game?

P.S. I use this idea in Red Orchestra game and the game engine allows to apply angles.

Yes, that happened in the AAR.

[WineCape]

... I'm curious, when you mention that you modify visual scanning for soft factors, do you take into account different optics and such. In other words, I've read many times that German optics were superior to U.S. optics because they used a more precise range finding method ...

BTW, CM simulates the "Zeiss factor" for German gun accuracy. This optic advantage is quite noticeable at greater ranges (800m+) in beta tests done recently - Sherman M4A3's vs PzIV(H)'s. The Germans have less of an advantage the closer the distance, as both German/USA tanks guns had a relative flat projectile trajectory at shorter ranges.

Charles recently indicated the M4A3 Sherman's gun is practically a "point-and-shoot" weapon for the gunner at close ranges (up to 310m) given the flat flight path of the shell, while the German KwK40 L/48 gun has this effect to roughly just under 500m distance, given documentary evidence/tests. So any given advantage the German optics might have in gun shooting accuracy, ceteris paribus, is in part/wholly negated within these respective ranges of these specific guns.

[YankeeDog]

I'm not talking about a "chance to hit", I'm talking about a "chance to see". The game has some method of determining what an AI unit can and cannot see. That's what I'm trying to determine. How do you simulate eyes scanning terrain?

I was responding to the first paragraph of your post, in which you made statements about optics and "chance to hit", which is why I quoted it. I did not answer your questions in the following paragraphs regarding spotting, because I don't understand enough about this aspect of the engine to give an accurate answer.

Steve has made statements that the game takes into account the actual location, facing and field of view of individual members of an AFV (see, for example, the stickied post "Some Bugs Are More Interesting Than Others"). But I'm not sure how exactly the engine calculates things like cover and concealment, spotting a moving target vs. a stationary one, etc.

[MikeyD]

Question, if a tank placed at an angle to it's opponent.....should create higher chances of bouncing off the armor. Does this effect apply to our game?

The angle of the vehicle, the vehicle's inclination on uneven ground, the trajectory of the shell, even whether the armor has been face-hardened and how brittle the projectile is taken into account. Every once in a while you catch a distance Panther sitting on a reverse slope. With some luck you get that prized 'hulltop penetration' as your shell's trajectory starts arching down.

[ClarkWGriswold]

I was responding to the first paragraph of your post, in which you made statements about optics and "chance to hit", which is why I quoted it. I did not answer your questions in the following paragraphs regarding spotting, because I don't understand enough about this aspect of the engine to give an accurate answer.

Understood. I should have stated that differently.

Steve has made statements that the game takes into account the actual location, facing and field of view of individual members of an AFV (see, for example, the stickied post "Some Bugs Are More Interesting Than Others"). But I'm not sure how exactly the engine calculates things like cover and concealment, spotting a moving target vs. a stationary one, etc.

I've read through that entire thread and understand the facing of each soldier and limitations on their field of view. What I'm really curious about is how the game simulates someone seeing and identifying an enemy. There's been a lot of talk in this thread about center of mass targeting versus targeting a specific area, but before anything can be fired upon it has to be spotted. I think that's where the real meat is for me, because in reality units were often extremely difficult to spot. I can imagine a tank back in a treeline being virtually invisible until it shoots (and perhaps even after that). And beyond being hard to spot, ranges were determined by knowing the size of the object being shot at and how large it was in the sights. So, not only do I find this interesting from a game mechanics perspective, but I think it's a key consideration when a discussion about targeting a turret comes about. Targeting a side-on tank in full view is one thing, but how much more difficult is it to get an accurate distance when all you can see is the turret (hull down)? And how much harder is it to see just the turret in the first place, before you even have to worry about determining distance?

I think these are important considerations, especially given that CMx2 doesn't use a probability model like CMx1 did (the very thing that people claimed gave weak turret vehicles a worse chance when hull down).

[WineCape]

An observation ...

"After-penetration effects, ranging from 'nothing' to armor spalling to brew-ups to casualties, seldom results ...... in a "one-shot-clean-kill" aka 1-shot-destroyed status for these specific tank duels [at 800m+]. Better get 2 guns bore sighted on the same AFV @ the same time (double overwatch!) in your QB if you want to kill your opponent quickly and outright. "

This was for the PzIV(H) + Sherman M4A3, head-on, non hull down, duels 800m+. Bets are off under 800m and/or side shots

[Battlefront.com]

There is a big difference between Spotting and Targeting. Spotting is more-or-less binary. Either you see something or you don't, while targeting is all over the place in terms of where the aim point can be. Now, spotting is a little more flexible than that because we have different levels of intel and there are now Sound Contacts too.

What happens is all of the factors I mentioned previously give the unit a chance of seeing the enemy (we do individual Soldiers, Crew weapons, and Vehicles). The stronger the results of the visual calculations, the greater the chance of spotting all the way up to max. But it generally first goes through one of the lower levels first. And it can progress up from there, or down depending on what later scans come up with.

Equipment in vehicles definitely is a part of spotting. The optics used for targeting are factored into the mix, but they are more important for targeting since they generally aren't great for spotting. The Commander being unbuttoned, for example, is BETTER than any other optical system within the tank. When buttoned things like periscopes, cupolas, vision blocks, etc. are all taking into consideration based on where the target is in relation to those optics. A forward looking periscope, for example, doesn't do a damned thing for the tank when something is directly behind it, but a cupola does. Blind spots are also simulated. For example, earlier Shermans had a rotating periscope for the Commander that only swings... er... 340 deg or something like that. It's not the full 360, whatever the case is. Which means there's a section of arc that the tank literally is blind to the tank commander (there's no cupola in that version).

Basically... what you expect to be simulated explicitly is. Including the Germans having some advantages in terms of range finding and distant targeting SPECIFICALLY and not getting some across the board "German über optics bonus". That would be very, very wrong since in some circumstances the German optics don't have an edge over the American type.

Steve

[LongLeftFlank]

There's something I never quite fully grasped (and I may not be the only one): I always understood Spotting to be based on a grid of Action Spots (to keep the computation load finite). So how would a unit within the Action Spot actually be Spotted? Or does it have its own "Spot". Or do I have my apples and oranges all hopelessly confused?

[Elmar Bijlsma]

A redux version:

Spotting from actionspot to actionspot is only the first coarse check to see if it is even theoretically possible. Only if that check is positive then LOS is checked more precisely on the 1:1 basis.

[LongLeftFlank]

Thanks -- that makes sense. And "1:1 basis" means the LOS is traced to the actual physical location of the vehicle or the "center" of the infantry unit (not to the individual pixeltruppen)?

[ChrisND]

Thanks -- that makes sense. And "1:1 basis" means the LOS is traced to the actual physical location of the vehicle or the "center" of the infantry unit (not to the individual pixeltruppen)?

Individual soldiers. You can see that now in CMSF.

[LongLeftFlank]

Ah, of course you're right. Even though I can't issue a Target command to hit an infantry unit without being able to draw a LOS line from my shooter's "center" to the target's center, my individual squaddies out on the sides will shoot at his men on their own initiative. It works very well, actually.

[Battlefront.com]

The easiest way to remember what an Action Spot does is to tell the system, relative to a specific unit trying to spot... "there's NO chance of anything being spotted here, so don't bother looking because even if there is something here you absolutely can't see it. However, there IS a chance of spotting here, so if there's an enemy unit here it needs to be checked for spotting."

Massive reduction in calculations.

Plus, Action Spots are the underpinning of the terrain mesh and terrain features. So it's all nice and bundled up together into one consistent grid.

Steve

[Kanonier Reichmann]

So, how long will it be do you think for the game to evovle to the point where the action spots and grid will be on a 5 metre square basis? Perhaps by the time your Eastern Front version of CMx2 is ready?

Regards

KR

[stikkypixie]

So, how long will it be do you think for the game to evovle to the point where the action spots and grid will be on a 5 metre square basis? Perhaps by the time your Eastern Front version of CMx2 is ready?

Regards

KR

Sure, just buy a pc that is roughly 4 times as powerful as you have now.

[Vark]

By the time the Eastern Front module is ready Moore's law predicts he will be able to do exactly that and even more!

[JonS]

Because it's been mentioned - and mangled - several times, below is the full quote from Andrew Wilson's "Flame Thrower" (originally 1956, Corgi edition 1973, page 47):

Next morning, when Wilson and Ward had reported to the Adjutant, Barrow told them the truth about the Crocodiles; the flame thrower was terrific, but the tank itself a death trap. There followed a little catechism about British and German tanks.

‘What do the Germans have most of?’

‘Panthers. The Panther can slice through a Churchill like butter from a mile away.’

‘And how does a Churchill get a Panther?’

‘It creeps up on it. When it reaches close quarters, the gunner tries to bounce a shot off the underside of the Panther’s gun mantlet. If he’s lucky, it goes through a piece of thin armour above the driver’s head.’

‘Has anybody done it?’

‘Yes. Davis in “C” squadron. He’s back with headquarters now, trying to recover his nerve.’

‘And how does a Churchill get a Tiger?’

‘It’s supposed to get within two hundred yards and put a shot through the periscope.’

‘Has anyone ever done it?’

‘No.’

[WineCape]

Because it's been mentioned - and mangled - several times, below is the full quote from Andrew Wilson's "Flame Thrower" (originally 1956, Corgi edition 1973, page 47):

Classic quote!

[sfhand]

The easiest way to remember what an Action Spot does is to tell the system, relative to a specific unit trying to spot... "there's NO chance of anything being spotted here, so don't bother looking because even if there is something here you absolutely can't see it. However, there IS a chance of spotting here, so if there's an enemy unit here it needs to be checked for spotting."

Massive reduction in calculations.

Plus, Action Spots are the underpinning of the terrain mesh and terrain features. So it's all nice and bundled up together into one consistent grid.

Steve

This got me thinking about hull down units. When we talk about seeing the center of an action spot, and now action spots being part of the terrain mesh and terrain features, I default to thinking one needs to be able to see the actual terrain mesh or feature to spot there when in fact I don't as evidenced by being able to target hull down units which are on action spots whose terrain mesh and features are out of sight of the targetting unit. (wow, talk about a run-on sentence) This leaves me more unsure of how it works than I thought I was before...

[Redwolf]

BTW, CM simulates the "Zeiss factor" for German gun accuracy. This optic advantage is quite noticeable at greater ranges (800m+) in beta tests done recently - Sherman M4A3's vs PzIV(H)'s. The Germans have less of an advantage the closer the distance, as both German/USA tanks guns had a relative flat projectile trajectory at shorter ranges.

Charles recently indicated the M4A3 Sherman's gun is practically a "point-and-shoot" weapon for the gunner at close ranges (up to 310m) given the flat flight path of the shell, while the German KwK40 L/48 gun has this effect to roughly just under 500m distance, given documentary evidence/tests. So any given advantage the German optics might have in gun shooting accuracy, ceteris paribus, is in part/wholly negated within these respective ranges of these specific guns.

Don't forget that high speed projectiles as such improve hit probability even with the same optics.

Mis-estimating that a target that is actually 2500m away is aimed at as if it was 3000m away does less damage to the hit probability as the initial projectile speed increases. A higher velocity gun with the same optics will hit more often. Because you lift you aim less with a higher velocity gun and if you do it wrong the unintended actual path might still hit.

This is one of the reason why the German 88mm gun was so feared early in the war. Not only does it penetrate what it hits, it has better optics and even if the range estimation is screwed up it is still more likely to hit.

[Redwolf]

By the time the Eastern Front module is ready Moore's law predicts he will be able to do exactly that and even more!

In a single-threaded codebase?

[Gromit]

In a single-threaded codebase?

**shrugs**

Anything is possible when you're a brain in jar of Steve-concocted nutrients...

In all seriousness:

I would imagine that one of these days there will come a point where Charles determines that the effort to change all the code necessary to take advantage of CPU multiple cores will actually make sense from an effort vs. payoff point-of-view.

We definitely aren't there yet... your guess is as good as anyones as to when that will happen.

[LongLeftFlank]

**shrugs**

Anything is possible when you're a brain in jar of Steve-concocted nutrients...

They make gooooood stuff in Vermont. Whenever we visit, we always come home loaded down with maple- cheese- and pig-derived products. And there's the King Arthur flour.... mmmmmm.