LOS/LOF and damage model clarification

[Guest Big Time Software]

Lokesa had asked a question in the "C3 Bashing" thread that will probably spark more discussion, so I am making it a new topic. Basically the question boiled down to "if the LOS is drawn to the center of the vehicle, why have all the armor stats in there since you only have one aim point?" A good question, and one that we should have clarified in the discussions the other day.

A shot will hit the side and angle that the shot vector intersects, not some center point. So if the target tank is at a 45deg angle (L/R) and 80deg angle (T/B) the shot will hit whatever part of the tank using this info plus all the ballistics and armor stuff. However, the 'shot vector' isn't _100%_ computed at 'launch'. It basically is, but the _exact_ part of the vehicle hit is done next, and is based on what percentage of the front, side, or rear is visible, etc. For example, a tank coming toward you at a 45 deg angle is presenting both its front and its side, so you could hit either one to the degree it is exposed.

In response to Thomm's suggestion of having each polygon looked at for determining hits...

We can not do this. First of all, the graphical representation of the tank does not always relate to the physical make up of the vehicle. It might *look* like it does, but we sometimes use tricks to keep the polygon count as low as possible. Remember that the sense of detail comes from the texture overlayed on top of a geometric shape. What looks like the turret storage bin on a tank might actually, so far as the 3D model is concerned, be a part of the turret. This means that if we defined that polygon to be turret, the storage bin would have the same armor value.

On top of all that, PE doesn't have to deal with the numbers and variety of vehicles we do. Doing the calculations to find the exact polygon is doable for something like 10 vehicles or less, but CM will regularly have more than this. Heck, I was just playing a game to see what the winter graphics looked like and, and without really thinking I had placed almost 20 vehicles in it. Knowing the armor nuts we have running around out there, I am sure this will be a paltry number to them

As with everything else in the game, the DATA for the vehicle has no relationship with the graphics what so ever. We could put in a box and call it a King Tiger and it would behave exactly the same as if we put in a 50,000 polygon model, complete with 3D shovels and tow eyes.

Even if we could afford to make a very accurate polygonal representation of each vehicle (got a super computer to run it on?) we would then have to specifically code each polygon for over 100 vehicles. I can tell you that this would take a VERY long time to do. Unlike the PE team, we don't have 10 people working on Combat Mission, not to mention a publisher's money to keep the wolves away. And even they don't have nearly this number of vehicles in PE.

Thomm also asked about our damage model. Like most everything else at the unit level, this is an approximation. Once the system determines that a tank is hit, various equations kick in to figure out what, if any, damage resulted from the hit. A hit might mean the main gun is knocked out, or perhaps a crew member gets killed, or maybe nothing at all happens, or the tank gets "brewed up", etc. This is all figured out using complex probability equations. We don't even attempt to see if a shot severs the electrical system freezing the turret, or knocks out a cooling fan thus overheating the engine. I know PE does a bunch of this stuff, but they are doing a tank sim with a very limited number of vehicles and little else in the way of combined arms modeling. And even their close attention to detail is an approximation, but obviously not as much as ours.

Short of it is that the extra nth degree of detail will buy only a smidge more realism than what we have now. But we don't think that anybody would notice it in action (i.e. do you REALLY need to know that one of your 10 vehicles' guns was damaged due to the optics getting cracked vs. the barrel getting dinged?), and even fewer would care because whatever differences it would produce would be very minor (i.e. both systems will likely report the gun being inoperable the same number of times from the same number of hits from the same conditions). The fact that it would delay the game and tax any current system means that this extra level is actually harmful. So obviously it isn't an option

Steve

[Herr Oberst]

We want Kursk! We want Kursk! We want Kursk!

Sorry, couldn't resist.

[Guest John Maragoudakis]

Awhile back BT suggested that beta testers would have an chance to create scenarios for CM. Then the best scenarios would be included in the game.

How about allowing players to create 3D models of a unit list suggested by BT. The best will be allowed into CM2 or CM,(if backwards compatible). BT could suggest the 3D package they prefer, the gamers go out and buy it on thier own. Gamers can learn 3D modeling through the process and BT will have many models to choose from, for free.

[This message has been edited by John Maragoudakis (edited 05-28-99).]

[Thomm]

Yes, I think I forgot that the tank graphics will usually not be drawn to scale, which brings up a visual issue: I assume that AT shells will have tracer characteristics. Does that mean that they will just go through the enlarged tank without leaving any trace (in the replay phase), if they do not hit the physical tank ?

I just realize that my assumptions on the shooting stuff were all wrong: Again it is statistics instead of vector algebra . Whatever.

<BLOCKQUOTE>quote:</font><HR> Even if we could afford to make a very accurate polygonal representation of each vehicle (got a super computer to run it on?) <HR></BLOCKQUOTE> I beg you pardon, but since you can move detailed 3D polygon models (yes, they ARE beautiful) around in real time at e.g. 15 f.p.s you should be able to do a shell hit calculation every let say 10 seconds. Who will notice ? It would happen in the turn evaluation phase, anyway ! And the armored storage bin would not bother me the least ! And what special code do you need for the polygons ? You just had to store a value for armour thickness of the polygon, and maybe what group it belongs to (turret, superstructure, hull, ...) for damage evaluation.

Okay you are right: Thou shallst not be immodest ! Lets wait for the supercomputers !

But: Do the kill probabilities change when the turret turns ?

[Thomm]

Sorry ! I did not want to behave like a spoiled child !

You know what: I wish you would put all your formulae on a webpage, one after the other, with all modifiers, standard deviations, everything and put an end to these discussions ! Oh, here goes the spoiled chikd again !

[Guest Big Time Software]

My point about the polygon count is that to make the kind of model which would allow a point for point representation of a tank, the way you described, would require a super computer alone. The hit calculations multiplied by potentially dozens of tanks at pretty much the same time could also be substantial. Both will chew up lots of development time. And what would this add to the game and simulation aspects of Combat Mission? Not much at all.

In terms of tying the data to the polygon, it isn't nearly as easy as you think it is. It would take weeks, if not a solid month, to write up a tool to do this. Doing it by hand would take just as long. It took about that long to code up a tool just for the game to understand how to map a texture to the model, which is far less complex then what you ask for. You must understand that a 3D program outputs nothing but coordinates and some related data. There is no way to define individual components and assign them sets of variables without coding up a tool or doing by hand. Neither is easy to do by a long shot. And again, what would this do for the game? And how could we do this without the more detailed models? The example I gave about the storage bin is but one. We have a tank that has the running gear, lower hull, and skirt armor using the same set of polygons. Without changing this to be more polygons the tracking of where each shot goes will be far less accurate than the system we have now. So this would actually harm the game.

The thing is that we can do all sorts of things to make Combat Mission a bit more realistic. But there is a point of diminishing return. We have probably already put in dozens of features, big and small, that fall into this category. But we are out of time for stuff like this, so we must remain focused on doing things like the AI, internet play, game balancing, and models that we know we can do and the game won't barf on if there are too many on the screen at once. We know what the limitations are out there and are dealing with them quite well in our opinion.

You can see each shell and you can see where it impacts the tank. The actual impact is at the center of the tank ( I think), but it all happens so damned fast that it is nearly impossible to tell. Especially if the tank explodes from the hit. You might not know that it hit just 2 inches left of the axle jack, which is next to the right rear storage bin, but who the heck cares? It hit the back of your tank, it is burning, nothing more is needed. The visual information we have tells you all you need to know. And sometimes even more! I just happened to pause the game with a shell in mid flight. I put the camera behind it, lined up its target, and hit Play and watched it rip apart a Stuart. Even the wife was impressed

Turret angle is, of course, factored into the equations. The turret is separate from the hull.

No to posting any of our equations. This would be the genre equivalent to finding the cure for cancer and uploading the DNA matrix (or whatever!) to the Internet. This is proprietary information in its truest form. Nobody needs to know how it works, especially because only a handful of customers would even understand it if they saw it. The results will speak louder than any equation ever could.

Steve

[Thomm]

Again very convincing ! Thanks for taking so much time ! Also good that you draw a line with regard to info disclosure. So that I know when to stop asking for more ! Actually I was close to asking for the source code

By the way: Why do you ALWAYS have to kill a poor STUART ? Even worse: in front of your wife, tss tss

Regards, Thomm

[Fionn]

20 vehicles?

Why oh WHY must you INSIST on modelling small (nay insignificant) unit actions..

I won't be happy till I have literal armoured phalanxes of Panthers over-running Moon's GIs ...

j/k but yeah ur right most of my scenarios will probably be armour heavy. I know its a character flaw but what the hell

P.s. My idea of a mixed force assault is using BOTH Super Heavy and Heavy tanks together. Eat that light infantry lovers hehe ,

[This message has been edited by Fionn (edited 05-28-99).]

[Fionn]

Thomm,

Something I learnt a while ago ... NEVER ask a programmer for their source code. They WILL attack with all the fury of the damned..

Oh and BTW I've also learnt that in one company I know the most heinous crime a writer can be accused of is asking for the source code .. (even if he didn't hehe)

Even I know the line has to be drawn somewhere as regards realism. As long as the EFFECT is realistic I can handle a few fudges in detail.

It's when the detail, the realism and the effects are ALL wrong that I get pissed but there's no worry of that with CM.

[Thomm]

Dear Fionn !

... comes from my academic background: Publish at all costs !

I also think that there is a huge difference between posting formulas, algorithms or even ideas on the one hand and to actually implement them ! How many competitors are out there who could abuse the mathematics, especially without the data to bring it to life ? Nobody can take the months of fine-tuning and debugging from BTS ! So why bother ? Seems that they are going to rule their niche for years to come, and the next game on the subject will for sure try to be "different" at all costs. So what harm would it do to distribute some distributions (haha) ?

Regards, A very naive Thomm

[Moon]

Alright, alright... Fionn... you wanted it this way: meet my "sticky bombs"...

[Guest Big Time Software]

Thomm,

We will make the armor penetration diagrams public, so you can see the way the algorithms work. We just won't show you the code.

See an example at http://www.bigtimesoftware.com/images/88_128.jpg

Charles

[Guest Lokesa]

Thanks Steve for clarifying the LOS/LOF, appreciate it. After posting I thought about the penetration charts and doh it started to make sense. Now lets see if I got it, a shot coming from the side uses a conglomeration of armor values for all the spots that the shell could possibly hit, these values are put together to make one value a % chance on damage from the side. So rather than trying to determine where exactly the shell hit and then doing the math for say turret mantlet, the turret mantlet is included in with all the other values from the angle of attack.

That seems fine to me, as you say for what the games purpose is, combined arms at platoon level, and seeing as how realistic results are truly what matters here, this approximating is totally acceptable.

Thomm, interesting point about rounds passing through vehicles due to scale. On the same line of thinking this would apply to inf as well as they are just an representation of the area that the inf are in and not representative of their exact locations. so it seems as though if we were to follow a shell along its flight path we might just get to see it go through a few soldiers heads a Stuart and a tree before finally impacting with it's target. Giving us the option of setting scale to our liking I'm sure this will happen to somebody. How big can we make our guys? there must be a limit, having gargantuan soldiers towering over the landscape would not do Maybe we can use scale as an intimidation tool online. WOW, that panther is huge!!! Look at the size of those grenades!!! ach, run away, run away I wonder what BTS has thought of regarding this. can anything be done about it?

[Guest Big Time Software]

Lokesa,

That's precisely incorrect. In fact we do just the opposite of what you say.

There seems to be a common misconception here about the way we do armor penetration, and I think it comes from "boardgame-think". You guys are thinking about how it would best be done with paper, dice, and some charts.

Specifically, people are asking how we arrive at a "kill percentage" and then do we roll "dice" to check against that percentage. As if we say "A PzIV can be killed from the front by a 76mm gun at 700 meters... let's see... 64% of the time. OK roll the percentile dice. Aha! A 57! I killed your Pz IV!"

Forget about all that.

I want to make this point crystal clear: Combat Mission does not compute or use any sort of generic kill percentage. We don't make a mushy average of various armor sections to arrive at a generic conglomerate armor strength. This concept does not exist in Combat Mission. Please erase it from your minds.

What we do is determine where the shell strikes, e.g. turret side, hull glacis plate, lower rear hull, etc. We do not figure it down to the square centimeter or individual polygon, but we do select a target plate that corresponds to the positioning of the target relative to the firer, and which plates are visible from that angle and to what degree. Then the armor of that plate and that plate only is considered in the resolution of the attack. If you fire at a tank from a 45 degree angle off from the front, you might hit the front or you might hit the side. You might hit the turret or you might hit the hull (upper or lower). And a shot hitting a turret front will generally behave quite differently from one hitting the hull side, as you might expect.

The colorful penetration diagrams we've posted are not based on some sort of generic averaging of target armor. They are the statistical results of "firing" thousands of simulated rounds at the given vehicle from all the angles and ranges on the diagram. Some rounds hit thick parts of the target, and others hit thinner parts, so that's why the likelihood of a kill is often neither 0% nor 100% but somewhere inbetween. A "kill percentage" can be deduced from the algorithms/diagrams, but the algorithms are not created from "kill percentages". You can look at the diagram and see that at a particular angle and range, the color is orange, which means about an 85% chance of penetration. But there is no d100 dice roll that's hoping to be under 85. Rather, from all the combinations of angles and target plates that can possibly be hit from that location and angle, 85% of such hits will penetrate the specific armor plate hit.

Here's a simple example. Say you're firing a gun at a Panther. The Panther is heading mostly towards you, but angled off at about 30 degrees. This means that, roughly, you can see equal silhouettes of the front and side of the tank. To keep things simple, let's say your gun can penetrate all the side plates (hull and turret) but none of the front plates. Since from this angle the front and side aspects are equally likely to be hit, this means you have a 50% chance of penetration. But nowhere in the program is there some dice roll of 50%. Rather, CM just determines (randomly, but based on aspect silhouettes and target angles) where the shell strikes, and then compares penetration power to the armor basis and that's what decides whether the shell penetrates.

Charles

[Mike D]

Great example Charles.

Now for a few more questions.

Let's stick w/ your example. Furthermore, let's say that it was a US 57mm AT gun firing at that Panther at a range of 1500m w/ whatever was a standard AT round for that gun. Assume the shot is determined to strike the quite thick and highly sloped front armour of the Panther. This, combined w/ the angle of incindence of 30 deg. as measured off the vehicle centerline I would most certainly guess would result in a non-penetration of the armour.

So nothing happens to the Panther right?

Q1:

Well let's see. Do you include a small probability that the shot might hit some soft area on the front armour. Say, an MG, or vision slot, penentration in the armour. If so can lower caliber guns that can't possibly penetrate the Panthers front armour still have a small chance for a penetrating shot?

Q2:

So the shot doesn't penetrate. What, if anything, does happen. Can the crew be rattled (esp. if they are green), stunned, demoralized, etc.??? If the shot in the above example doesn't penetrate and does hit the front MG mount and also doesn't penetrate is there a chance that at least the MG is knocked out?

Q3:

Let's change things up a bit and say that the shot was from a 90mm gun at 500m instead and that shot strikes the Panther squarly on the side of the vehicle (i.e. angle of incidence relative to vehicle c.l. is 90 deg.) and penetrates the lower hull. What and how does the CM game engine now calculate in order to determine the damage inflicted by the penetrating shot?

Hope everyone is having a great Memorial Day.

Regards,

Mike D

aka Mikester

[This message has been edited by Mike D (edited 05-30-99).]

[Guest Big Time Software]

Mike,

#1: Yes, we call those "critical hits" and the armor is reduced by about 40% in strength before testing penetration. But these are fairly rare (1% or less of the hits).

#2: We don't rattle the crew (I felt this would be too easy - a "gamey" way - to scare off King tigers). However, shots that hit the vehicle tracks or external gun or fire system (optics, etc.) are not required to do the armor penetration check, and can immobilize or damage the main gun. So a Panther or King Tiger is at least somewhat vulnerable, even from the front. If enough Shermans shoot at that KT, one of them is going to blow off a track or take out the optics or even strike the gun barrel itself. Then the KT crew won't be happy campers anymore. In fact, that's often how real life Shermans had to deal with heavy panzers - just sheer weight of numbers until they could "dent" something important. Then move for a flank shot to kill.

#3: Factors involved are:

- Size of shot

- Type of shot (more explosive the better)

- Size and weight of vehicle

So a big shell like an 88 will almost certainly destroy a Stuart. On the other hand, a tungsten round from a 57mm is not assured of destroying a Panther (small, subcaliber round with no explosive charge against big vehicle). Of course, when I say it has a "low" chance of kill from a penetration, I mean something on the order of 50-60% (off the top of my head) as opposed to the 95+% of the 88mm versus the Stuart.

Crew casualties are also considered, and even if the tank sustains no significant physical damage, if a crew member or two becomes a casualty, the crew will bail out. (Sometimes elite crews won't bail out after only one casualty).

Charles

[jkeen]

In a few of the books I've read about World War 2 (Against the Panzers is the only one I can think of right now), it states the veteran gun crews would attempt to aim for vulnerable points on the tanks. One example was fighting against panthers in the fog, the gun crew would aim for the muzzel flashes of the machine guns. Does Combat Mission improve the odds for a critical hit based on how experienced the gun crew is?

[Guest Scott Clinton]

#3: Factors involved are:

- Size of shot

- Type of shot (more explosive the better)

- Size and weight of vehicle

This may have been addressed before but...

Do you model different Ammo types? AP, HVAP, APHE, etc...there can be a WHOPPING BIG difference in the penetration in the TYPE of round used.

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

The Grumbling Grognard

[This message has been edited by Scott Clinton (edited 05-30-99).]

[Guest Big Time Software]

Yes, the chance of an experienced gun crew hitting a weak spot is increased by virtue of them simply having a greater chance of hitting in the first place. (BTW, I have Against the Panzers and found it to be a tad biased towards US forces in places)

Yes, shot types are simulated very accurately for exactly the reason you mention. However, the availability of certain rounds depends on weapon type, nationality, and date. Something like a tungsten round will be rare for a German crew to have, especially late in the war, due to an acute shortage of the metal.

You should also note that different types of armor are also simulated to complement the different shot types.

Steve