TIGER SIGHT--SIMPLIFIED? Damage model--Bogus?

[Txema]

ARMOR MODELING in STEEL FURY

I have read the previous posts and I have seen that probably you are not understanding properly how Steel Fury models the armor protection of the tanks. I am going to try to explain what I understand about this issue.

Steel Fury attempts to simulate most of the relevant issues in tank warfare with very high detail. Armor protection is no exception and is also simulated in detail.

To do it, the "armor_thick" entry in the .engcfg file is used as the "base" armor value for every vehicle. In the case of the Panther it is defined in the "Panther.engcfg" file as: armor_thick=100. This means that the "base" armor thickness in the Panther tank will be 100 mm. The actual armor thickness used by the simulator in each surface of the tank is defined in a tga file (panther_armor.tga in the case of the Panther). This file has colors that follow a grey scale, going from black to white. The "RGB" values corresponding to those colors can go from zero (totally black) to 255 (white). The actual armor thickness used by the simulator for the corresponding surface is calculated as: RGBvalue%*armor_thick. For example, the Panther has an upper hull front armor equal to 85% * 100 = 85 mm, a lower hull front armor equal to 65% * 100 = 65 mm, and a side hull armor equal to 40% * 100 = 40 mm. The front turret mantlet has a thickness 105%* 100 = 105 mm, and so on. As it can be seen the armor thickness of the tank is coded in a very datailed way.

To understand how the grey armor surfaces present in the panther_armor.tga are applied to the tank, it is very usefull to see the texture file used to paint the tank ("techn_panther_c.dds" in the panter case). Actually the textures are applied to the tank in the same way as the armor surfaces.

All of these armor values represent the real thickness of the armor plates. The effect of the sloping of the armor plates is taken into account in the game engine when it calculates the angle of impact of the shell, and reduces the penetration abilities of the shell accordingly. Therefore the armor thickness coded in the models shoud be the real thickness, not an effective thickness to take into account the sloping of the armor (!).


P.S. The "panther_armor.tga" file can be found in the "data\k42\loc_rus\armor_maps" folder
       The "techn_panther_c.dds"  file can be found in the "data\k42\loc_rus\textures\techn\tanks\heavy" folder


Txema

Edit: updated to incorporate de corrections suggested by Kyth

[Kyth]

Hi Txema,

Thanks for the explanation about the armor values, and it's much appreciated! 
I agree with most of what you're saying, about the armor map tga files and their use as a 'skin' to determine armor thickness.

But, pardon me, I have a slight quibble about the grey scale, and it's relating to the "RGB" (red, green, blue) values mentioned earlier in the thread.
As I see it, in the armor-map tga files, the RGB values can range from zero (totally black) to 255 (white).

Example:
For the Ferdinand, I can confirm that the shade of grey going over the front armor plates has a RGB value of 200.
Hence, the "armor_thick" value of 100 (from the engcfg file) , multiplied 200% (the RGB value), gets a result of 200mm for the frontal plates.

Also:
The side armor of the superstructure's shade of grey has a RGB value of 82, so,
100 * 82% = a side armor value of 82mm. (Near enough, I think, to the real-life thickness of 80mm)

The basic values are actually quite correct! When it comes to details, however...
I think I need to do a separate post, at a later time, 

[Txema]

But, pardon me, I have a slight quibble about the grey scale, and it's relating to the "RGB" (red, green, blue) values mentioned earlier in the thread.
As I see it, in the armor-map tga files, the RGB values can range from zero (totally black) to 255 (white).

Example:
For the Ferdinand, I can confirm that the shade of grey going over the front armor plates has a RGB value of 200.
Hence, the "armor_thick" value of 100 (from the engcfg file) , multiplied 200% (the RGB value), gets a result of 200mm for the frontal plates.

Hi Kyth,

I have checked it and now I do agree with you. My post was certainly wrong in that regard... Thank you very much for your correction !!!

I am going to update my post with your information.


Thanks again !!!


Txema

[Kyth]

Hi Txema,

It's okay,

I actually have a few things I need to ask, which puzzle me a bit:

The engcfg file for the Ferdinand states the "armor_thick" value as 100. Is there a limitation for this value in the file? Is 100 the maximum allowable? For instance, would "200" be recognized?

The same question goes for the other information such as arm_fwd, arm_side, arm_back and arm_up. (I assume these are for AI consideration.) Are they also limited to 100?

By the way, looking at some of the armor maps on the 3D models, it strikes me that it may be causing some of the weird AI behaviour I've observed. The most grievous example being, the PzIII! It's still a hunch, though. 

[Kyth]

ADVISORY--I don't know if it is related or not, it could be something else screwed up, but since I installed this (my own mod) I am having a problem with my tanks shedding both of their tracks for no apparent reason. Need to figure this out.

Edit: Uninstalled and reinstalled all my mods--haven't noticed the problem. I don't think it was caused by the LtVehTweaks mod--don't see how it could have been.

Have a new version of the mod--think it works a little better. Armor strength value is left at 1800 for the Half-Tracks, but armor Quality is reduced from 0.9 to 0.5. So far, they resist machine gun bullets, but are knocked out rapidly by main gun rounds. Link to new version below.

2cnd Version:
http://www.mediafire.com/download.php?kpapbanz5dx9pjz

1st Version:
http://www.mediafire.com/download.php?dz0badqhp5eawi4

Here is the link to the mod I made to tweak the light vehicles. To install the mod, place it in your MODS folder and enable it with JSGME. This mod will only work with SPM 1.5 Beta and should be enabled subsequent to (after) SPM 1.5 Beta and after all Patches and Fixes for that mod--there should be no problems, but if there are please alert me and I will try correct them. Also, your comments and a reactions to the changes I have made are very welcome, and any advice as to other values which can be altered to improve the realism and playability of the game is very much welcome as well.

Edit: after playing a bit, I think the German Half-Tracks are still too tough. Dammit, a half-track should not take a direct hit from a HE round to the engine and keep going, gotta fix it...

The goal, as I see it, is accuracy, fairness and consistency for all players, so that we can have a common experience and a common challenge with this sim. I'm happy to say this group, so far, doesn't strike me as the type that would tweak files for the purpose of making things easier (or even overly tough) on themselves.

   Yeah, I think that for at least the lion's share of us, the over-riding goal of making changes to the status-quo is to increase realism, whether that makes the game harder, easier, or whatever. You could say that my tweaks to the light vehicles makes the game easier, since now you can engage trucks more effectively with the machine gun. All I can say is, Real Life is in this case easier than the game was. Look on YouTube at American troops engaging technicals (militarized commercial vehicles). They just blast 'em with machine guns, never main gun rounds. Heavy machine guns (M2), light machine guns (M240), assault rifles, sniper rifles, whatever. A .45 or a 9 mm is a viable weapon against a truck, car, or motorcycle.
   Anyone who wants to is free to make the game as easy as they like--just make your tank invulnerable, or give it infinite ammo, but I don't think they will find many (if any) like-minded people playing the game. Who would take the time to slog through the twisted verbiage of the manual to learn the difference between OF350 and F350 and then turn the game into an arcade experience? So those of us who "master the game" are very likely to have identical wants in terms of realism.
   From my initial delving into the cfg files, it appears to me that there are likely NUMEROUS errors in the details of the vehicles. In many cases, values seem to have been grafted on to them out of sheer laziness. Why did the motorcycle weigh almost two tons? Because it was classified as a car, and that is the value that was being pasted in for cars and trucks. No matter that it made a motorcycle hit by a heavy shell behave like a truck. Perhaps we were not expected to notice. I do wonder about Engine Power values. I have not looked closely. Now, after a brief check, I see that motorcycles have an engine power rating of 75 and trucks have an engine power rating of 40. (?) That explains how a motorcycle that weighs 1.8 tons could move, and move faster than a truck. This will probably require some experiment and tweaking as well. As we come to understand how these values relate to the trucks, etc, perhaps some refinement can be made to the tanks as well to correct some of the strange behavior. I would very much like to find out where the value for setting turret rotation speed is. The Tiger I and the KV-2 are prime examples of tanks that suped-up turret rotation which has a very negative effect on the tactical realism possible in the game.

Hi Nodlew,

About the new tech-cfg files, which ones were changed? Was it just the trucks and half-tracks, or the tanks also? I'll be giving them a try soon,

Thanks,

[nodlew]

All right, Frinik, consider me pacified. Maybe I'm getting thin-skinned in my old age--or maybe I have PTSD from too many arguments on the internet over the years. I think my first conversation in a chat room on AOL was probably an arugument. Ah. Those were the days.

Kyth, the changes were to the trucks and half-tracks, and the "mod" needs to be reworked. At present the changes are made to all of the Sdfkz Half-tracks (armor quality reduced from 0.9 to 0.5) and the trucks and cars--armor strength reduced to 1000 (from 1800) and mass on the Opels and the Zis5 trucks increased to around 2 tons to correct some goofiness. Mass on the BMW R12 reduced from 1800 to 500. So now the mod consists of a hodgepodge of changes, but the effect of these changes strikes me as better than stock. Half-tracks are more easily damaged by main gun rounds (although they several rounds can still be necessary) but invulnerable to machine guns, and trucks, cars, and motorcycles are vulnerable to anything with bullets in it.

As for the Elefant I am torn between giving it 200 mm front armour or splitting the value in half between 200mm upper front and 100mm lower front although the sloping effect actually added another 100 mm effective?

What I would like to do with the Elefant is play with reducing its engine power. I'll let you figure out the armor values, and adopt your modifications when they arrive. But for me, the most irritating thing about the Elefant, as is, is its nimbleness. I want to see it lumbering across the battlefield and slowly rotating on its tracks to train its gun, not zipping around like a dune buggy. Not spinning in place to keep its gun trained on a T-34 (me) running parallel to it at top speed.

[frinik]

No worries Nodlew only an exchnage of point of view as far as I am concerned.

|txema I am aware of the armour tga files but as far as what I have observed the values contained in the tech-cfg files seems ot have a greater impact on the actual resuls obtained in the game. I did a little experiment increasing the armour thickness value of the Tiger II to 120 mm and the front armour to 120 mm( a mix between the hull value of 160 mm for the Henschel turret and 80 mm minimum thickness of the lower front armour = 240 div by 2).Not an ideal calculation but the result was quite interesting; whereas before even 85 mm shells used to kill my Tiger II at 1500 metres, punching it's thick sloping armour ( highly unlikey considering war and battlefield reports.|Even the Soviets themselves advised their T34/85 tank crews not to attack a Tiger II unless they were at 800 metres or less and then try to hit him on the thinner side armour or in the engine) now only JS 2 or SU122/152 shells can actually kill my Tiger II at distances of 1200 metres and beyond which matches battlefield reports or observations from both sides.The armour tga files may actually matter more in terms of visual damages that you can observe on the tanks post battle.

\It does not also explain why I found 3 different values for the T34/85 frontal armour thickness 75 mm and 90 mm between mod 1.5 unoff and beta 1.5.Nor why has the Tiger I been ascribed 3 different frontal armour values in 3 different mods\91.4, unoffcial 1.5 and beta 1.5).

I also quesiton ascribing a set value of 100 mm to tanks as different as the Tiger I, TIger II, Panther?

KYTH to answer your question 100 mm armour thickness seems to be the max value ascribed to heavy tanks but by no means can't you increase it.As I explained above with the Tiger II I jacked it up to 120 with observable effects.Likewise I am wondering why has the T34/76  on the Soviet side been given armour thickness of 52 mm while it's exact German counterpart ,the captured T347r, is only given 47 mm?? Nor do I understand why is the T34/85 given a value of 90 mm ALTHOUGH DESPITE IT'S SLOPING ARMOUR IT'S REAL ARMOUR THICKNESS WAS MUCH LOWER THAN THE TIGER II ALSO SET AT 100 MM? I don't buy those values as they seem to indicate a sort of tendency to balance the game in favour of Soviet armour in the game at the expense of realism....

Which is why I am grateful the mission editor and game file folders are so accessible to change and modifications.

|cheers

[Stig]

Frinik wrote:

I don't buy those values as they seem to indicate a sort of tendency to balance the game in favour of Soviet armour in the game at the expense of realism....

Which is why I am grateful the mission editor and game file folders are so accessible to change and modifications.

+1000

I too am alarmed at the tendency of games coming from Russia to have political bias in them. Starting with IL-2 and emerging in nearly every Russian-authored game I've ever heard tell of. But of course, there are any number of American-authored games with a severe case of "John Wayne-itis" in them, as well. And, of course, there's always the possibility of plain, old fashioned error. But what's more important is the ability to be able to FIX error/bias when it's identified.

IMO, games with "open" files are all but a must (at a minimum Read Only, where one can SEE for oneself what values are used, and not have to take any developer's word for it. What came to be a bad joke in the IL-2 community was the lead designer, Oleg Maddox, and his tendency to answer charges of inaccuracy and bias, or deflect questions on where his numbers came from with, "Is correct, be sure."

At the minimum, we players should be able to see the numbers, understand them, and be able to carry on an intelligent discussion on accuracy. Developers who put forth incredible excuses like, "We paid a LOT of money for our statistical data, and we won't share it with anyone" or dodges like that, lose any credibility whatsoever.