Jump to content

Why Do IS-1s Back Away From German Heavies?


Recommended Posts

I can understand slow ROF vehicles (IS-2) backing away from an armor fight, even when the gun (122mm) has the potential to penetrate the big cats. What I don't understand though, is why the 85mm IS-1 is also timid when it comes to the big cats? A Panther will cause the heavily armored IS-1 to back away. Sometimes this 'retreat' will occur before the IS-1 even fires a single round. Is there a good explanation for this behaviour?

Treeburst155 out.

Link to comment
Share on other sites

I'd guess this behaviour has something to do with a flawed AI: it somehow calculates the hit probability of each consecutive round fired, coupled with the probability that the hit will knock the target out.

Since the Panther has better accuracy, a ROF equal to the IS-1 at least, and a more potent round, the IS-1 backs off.

[EDIT]

The odd part in this theory is that German tanks rarely behave in this manner. Perhaps it's due to nearly universally better performance of the German guns, and the fact that just about all Soviet tanks are vulnerable to the vanilla 75L/48 gun.

[ March 27, 2004, 06:24 PM: Message edited by: Bone_Vulture ]

Link to comment
Share on other sites

Untermenschen cannot possible be expected to stand up to the valiant men of the panzer waffe. It says so right here in Signal, so it must be true. Why anybody bothers to look further for explanations of such things, at this point, is the only real mystery - since there are about a hundred such cases in CMBB.

"Russian heavy tanks weren't meant to fight tanks". So, do all German tanks cower in 1941 and 1942, and all US ones in 1942 and 1943? Because they had the same doctrine. Hmm, they don't. It is only when there is a class of tanks comparable in combat power to the better German ones, that such slavish doctrinal determinism becomes a subroutine, and actually neuters dueling performance.

[ March 27, 2004, 09:56 PM: Message edited by: JasonC ]

Link to comment
Share on other sites

The thing is also dependent on the position in the terrain. The hit and kill chances vary with much more than just hulldown or not and the retreat trigger depends on this.

But in CMBB it was usually the Russians drawing the short end, one of the reasons why I gave up on it. But I remember that somebody verified that a Panzer IV lang chickens out when a SU-100 targets it.

Link to comment
Share on other sites

I hate tank cower behavior generally. I think a tac AI routine should override player orders as an SOP, only if it is going to be right 90% of the time. And present tank cower behavior is wrong more like 90% of the time.

It winds up hurting the Russians more, because of how they have to use their tanks, their particular weaknesses and strengths compared to typical German opponents, etc. Also because the Germans typically have smoke to pop when they do it, while the Russians typically don't. It is all a royal pain.

I prefer fights dominated by infantry purely because of this issue. When any armor is present, I prefer nothing with even one plate impenetrable to typical full sized enemy AT weapons (not specialty weapons, but common ones).

Epic tank battles on the eastern front would be one main attraction of CMBB. For me, there is no such attraction because I do not even attempt to simulate them. The results in practice are such Signal magazine cartoons, for all the wrong reasons, that I can't watch 5 minutes of them without bursting out laughing. Regardless of which side I am playing.

I really think tank cower should just be gutted as it is right now. I've suggested a more subtle reform of it in the past (only if targeted, only with no round in the tube, only if chance of killing the enemy yourself reads "rare" or "none", only if being faced to within 45 degrees) - but that might be more work. I'd settle for it just never happening.

One man's opinions...

Link to comment
Share on other sites

Hear hear.

CM in general but CMBB in particular would be a much better game if the "upper half" of TacAI self-inflicted commands would just be replaced with "stop and go to cover (if infantry) and shoot back" instead of trying to get funky with non-working algorithms.

With "upper half of TacAI self-inflicted commands" I mean the stuff that is given and overrides player orders although the units is not yet paniced or broken. If my men run to the friendly made edge if they are broken - fine. If they crawl, not even paniced, out of a foxhole towards visible enemies in the same terrain type that the foxhole was in then it is just a misfeature and staying put would be more realistic overall.

I would so have loved the Russian front in a CM game but it was just too much. CMAK is better.

Link to comment
Share on other sites

What does me more anoys beside the slavic "Untermenschen" behavior...is, that my Panthers need 20 rounds (on longer Distances) to hit those for- and backward cowardycing T-Somethings, where it can hit and kill normaly with 3-4 rounds if it stay still on one place.

What does me too wonder, that those WWII Tanks start from 0-40 like a Ferrari....and with the real turning radius from the different tanks, the game would look a lot more different.

Link to comment
Share on other sites

One request.Would someone please set up a scenario editor test in which *any* russian AFV(as long as it is capable of knocking out a Panther from the side/rear)rolls up from a defilade position,and while in a hull down position,is able to fire at a buttoned up Panther that is facing away from the Russian AFV,while not hull down to the Russian AFV itself.

I simply want to know what happens,as I am not proficient enough with the editor.

Link to comment
Share on other sites

Originally posted by Bruceov:

Did tank cower behavior occur in real life?

Cower might be a wrong word.

IRL tanks would probably use shoot n scoot when they were within each others 'kill-envelope'. Just as riflemen will not stay exposed when the volume of incoming fire gets 'hairy', tanks do not stand still and 'take-it'.

I doubt even impregnable tanks looked kindly on being struck by non-penetrating hits.

The player and AI should use shoot n scoot if they desire more shooting before scooting.

The game should model backing away to take into account if the threat is targeting them. Lets say 3 JS tanks approach the side of a lone Tiger (that is, they only see one Tiger). The Tiger has its turret swung elsewhere. The JS tanks should not just back away. They might stand fast and shoot till the Tiger actually targets them

Link to comment
Share on other sites

Tittles is right about the basic issue. The problem is largely that cower in CM is a pairwise function, that does not take into account enough relevant battlefield information. If X makes A cower, it can also make B, C, D, ... cower, to the point where a horde can run from one. In reality, tankers could and did judge whether odds, distraction, immediate possibilities of maneuver and cover, etc made it safe to proceed.

So can CM players. The tac AI should not wrench things out of the players hands when they know such things. Players will keep vulnerable tanks out of view on their own, when conditions are unfavorable, because they don't want to lose their tanks to no purpose. When a player decides it is the right occasion to risk a weaker tank against a stronger one, it is usually a sound decision. He knows there will be help, another shooter, a flanker or hail fire by a large force, etc.

If the behavior had extra "flags" set that prevented cower e.g. when the enemy tank is facing the wrong way, when not targeted, when the would-be cowerer has a round ready - then such things might be detected by the routine. But right now it acts as though it is the only tank o the battlefield, and when facing an individually superior or even just a dangerous opponent, it should run away. The result is comedy, not realism. A platoon of IS-2s will cower from a lone Pz IV.

Link to comment
Share on other sites

Well, I have just tried this in a number of varieties, and I simply can not see the problem.

IS-1 is always hunting across a crest, regular, commanding vehicle (out of LOS) has a morale bonus, vehicular cover arc over the frontal arc in action. Panther is conscript, padlocked, buttoned.

1) Panther at 360m, presenting broadside to IS-1,

Panther dies, no cowering

2) Panther at 650m, presenting rear to IS-1

Panther dies, no cowering

3) Panther at 650m, presenting front to IS-1

Panther dies, no cowering

4) Panther at 1,000m, presenting front to IS-1, but no AP

Panther dies

5) Panther at 1,000m, presenting front to IS-1, with AP

a) IS-1 hunts, fires, upon return fire runs away

B) IS-1 hunts, sees Panther, fires, runs away

c) IS-1 hunts, sees Panther, runs away until backing on wood (later removed), then fights it out (i.e. rather tries to fight instead of presenting side armour to Panther) - broken off with no result because the wood skewed the result.

d) IS-1 hunts, sees Panther, runs away, at end of turn (still in LOS) reverse order is cancelled, 20sec pause order given, IS-1 starts ding-dong fight, sticks it out despite partial penetrations and crew casualty. IS-1 dies.

e) Commanding vehicle IS-1 now gets in on the action. Acquires Panther at 800m, engages, hits with glancing shot. Panther returns fire, IS-1 dies.

Conclusion - I think there is a whole gamut of outcomes. Treeburst's post setting this off is a pretty useless rant of the 'My Tiger just died wawawa' type, because we know nothing about the conditions of the engagement.

What I have just seen seemed very realistic to me - sometimes they fight, sometimes they don't. At the longer ranges, the Panther clearly has an edge, due to higher hit probability, and better penetration ability - the IS-1 only manages front turret partial penetrations (most engagements are HD in this setup), while the Panther can penetrate the front of the turret and the glacis at least partially, sometimes fully. The two times the IS-1 stood to fight it out under these odds, it died.

I am sorry for bringing an actual playtest into this discussion, you can now return to JasonC's regularly scheduled lecture on how BFC is shafting the Soviets.

Link to comment
Share on other sites

Actually, Andreas, the incident that triggered this topic was that I can't get my IS-1 to die. He keeps backing away from the fight. :D

I've always accepted Soviet armor behaviour with the rationalization that the slow ROF of the 122mm and the easily penetrable armor of T-34s justified the behaviour. Then I finally found myself with a nice, beefy IS-1 that did not have a slow ROF. I thought to myself, "Now this baby will fight!!", but it doesn't. smile.gif

BFC have made several design decisions that I disagree with. I just started this thread to see if I'd found another one. My ranting is confined only to the Molotovs, which I REALLY don't like.

Having said that, I've enjoyed CM for several years, with no end in sight. These design decisions I disagree with can't be that major if this is true. It's a good thing I can remove molotovs in the editor though. :D

Treeburst155 out.

Link to comment
Share on other sites

I was going to spend some time today reproducing what I consider to be undesireable Soviet armor behaviour. Then it occurred to me, what I consider undesireable is probably perfectly fine behaviour to BFC and their testers. It's all a matter of opinion. I don't think I can show bad enough Soviet armor behaviour to change anyone's opinion on that behaviour. If true, the current behaviour must be accepted, just like I must accept the existence of politcal views that differ from my own.

Treeburst155 out.

Link to comment
Share on other sites

Treeburst - I suspect that part of the reason mine dies is that it is commanded by me. It seems to happen to a lot of tanks.

If you want to go ahead and test it, I would be interested in what you find. Won't change anything for CMBB one way or the other, of course, since it will no longer be patched. When it comes to opinions on the matter - I am in favour of variability, unpredictability, and a coded-in attempt of my pixel soldiers to try and survive despite the orders I give them. I must say that my IS-1 did very badly in their attempt to do so.

Clearly what I have done is not an exhaustive test, but it also shows that the idea that this cowering is standard behaviour is incorrect. I managed to convince my IS-1 to fight. It was smarter than me though, but still died.

Link to comment
Share on other sites

2 IS-2s (mid 1944 model) vs. 1 Pz IV H. Pz IV H kill chances read "very low" to "none" at 730 to 1100m with various side angles. IS-2 kill chances read "excellent to "very good". Two outings.

First time an IS enters LOS, the Pz IV acquires and fires. The IS-2 halts, acquires, then reverses before firing, shoots on the move, misses by a mile. The other IS-2 enters LOS, halts, fires, continues forward after missing. The Pz IV hits several times but gets UFH ricochets. 2nd turn both IS-2s hunt into LOS, Pz IV stands, hits and bounces, the IS-2s each miss once, one fires a second time and kills the Pz IV.

Second outing, less cover and longer ranges, engagement starts at 1100m. Pz IV with kill chance "none" does not cower, fires repeatedly, bouncing hits off one IS-2. First IS-2 with LOS fires and misses, second hunts into LOS, halts and fires, first fires again and KOs Pz IV.

Of 6 chances, an IS-2 cowered once, missing its shot because of firing while reversing. The completely overmatched Pz IV never cowered.

Same situation but with a Panther in place of the Pz IV. Now the kill chances read "fair" to "OK" for both sides. A stationary IS-2 that starts in LOS, no hunt no arc no orders, reverses instantly, misses its 3% shots while moving, turns its side armor broadside, and is KOed by the 3rd shot, of the Panther that instead acquires instantly and fires until the IS-2 is dead.

The other IS-2 fast moves into LOS, instead of hunting. The instant it gets LOS (around an intervening house midway between) it reverses without firing. Next it is given a VCA. The Panther fast moves into LOS also with a VCA. The Panther completes its move and fires. The IS-2 begins reversing before firing a single shot, misses, and is KOed by the Panther's third shot.

Pz IV with kill chance none - cowered never.

IS-2 with kill chance excellent - cowered 1/6.

Panther with kill chance OK - cowered never.

IS-2 with kill chance OK - cowered every time.

Link to comment
Share on other sites

Originally posted by JasonC:

But right now it acts as though it is the only tank o the battlefield, and when facing an individually superior or even just a dangerous opponent, it should run away. The result is comedy, not realism.

I've got to disagree with that. Each tanker IS pretty much alone on the battlefield; from his own perspective, anyway. He's getting orders over the (bad) radio and relying on a pre-mission briefing.

Comrade Captain: "Comrade Sergeant! We need you to keep that Tiger busy with your T-34 while we send the JS-IIs to flank it. Don't worry, they'll show up in time to give you the help you need."

Comrade Sergeant: "Yes sir." (Under breath) "Suuuuure they will show up in time..."

I really think the way tanks react is very realistic. Tank crews - like infantry - aren't robots. They're people in big, ungainly monsters, fighting relatively blind. The ONLY way to stay alive in a tank is to be extremely cautious. Game players are omniscient, tank crews are not. You never know if that infantry you saw is screening a Pakfront, or if that enemy tank is alone or just covering for the rest of it's platoon's flanking maneuver – and if you don’t retreat, you’ll be flanked and killed.

As it is, a real tank platoon commander would have nowhere near the control over his platoon that a CMBB player does - and you want more control?

When I was a tank commander (M-60A-1s, later M-1A1s) in real life I was happy if I could keep radio communication with my section leader and show up at the same objective. To expect to-the-second command reactions and the sort of clairvoyance and/or foolhardy bravery required to charge into a dangerous enemy would have been absurd in real life - and I had years of training and much better tanks and equipment.

Also, we don’t know all of the variables at work within our simulation.

Consider this little incident, from Hans von Luck’s book, “Panzer Commander.” In 1945, Luck has met with a former classmate, who is commanding a section of Panthers. He’s going to move his section into contact when von Luck warns him:

“For God’s sake, stay here, or steal forward on foot. There are Stalin tanks dug in, with nothing but the turret showing; they’re dangerous.”

Pipkorn didn’t listen to me, but climbed back onto his tank and drove out of the woods. A few moments later I heard heavy tank fire and Pipkorn’s tanks came rolling back to me, into cover: beside the gun lay the body of my friend Ruediger, who had to die now, so shortly before the end.”

So, here we have a section of Panthers. They move up, take what seems to be a single casualty to a tank-riding officer – probably a non-penetrating hit – and the whole section of Panthers retreats from the Stalin tanks.

Would you decry the game as “unrealistic” if this happened?

Omniscient, super-brave tanks might make for a more fun game, but it won't be a more realistic simulation…

[ March 28, 2004, 05:07 PM: Message edited by: Cessna ]

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Unfortunately, your content contains terms that we do not allow. Please edit your content to remove the highlighted words below.
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...