Tanks Falling Off Bridges-What, No Damage??

[Duke of Earl]

Bonjour fellas,

Why is it that when a tank falls off a bridge... like a Russian T-34 during a winter battle... it suffers no damage or crew casualties? Very good shocks? Really fluffy snow? hahahahahahaha

I did a search for this topic in the CMBB threads, but came up empty...

THE PAW OF THE TIGER

SPOILER ALERT - SPOILER ALERT - SPOILER ALERT

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

Downloaded this scenario from the Depot... played it as Axis against the Russian AI... Allied attack during frozen winter... heavy pine forest cover with 2 small villages... basically, I must defend with a 150mm spotter, 2 Tigers and 3 Pz IV's on a ridge against 30 or so T-34's and KV-1's... object is for Russians to exit the map behind me and screw up the Leningrad offensive...

Anyway... Turn 1... laid down a preplanned 150mm barrage on a key crossroads feeding 2 stone bridges... created one helluva traffic jam for the AI... which, of course, tried to funnel most everything across only 1 of the 2 available bridges...

So... scored several hits on the road surface of the main bridge... no apparent damage... scored a direct hit on the top of 1 T-34 at entrance to the bridge... instant obliteration... then, the barrage immobilized or destroyed several T-34's on the main bridge itself... very tight squeeze... as I intended... AI had to push them off the bridge...

But, during this process... 3 operational T-34's also fell off the bridge into the gully below... I watched them do it... the gully had steep slopes and was blocked at both ends... fortunately for me...

Because, the T-34's suffered no damage from the fall... and it was quite a drop too... no immobilization... no gun damage... no crew casualties... nuthin...

Those bad boys kept moving around in the gully trying to get out... but they couldn't handle the slopes and the ends of the gully were plugged up with pine forests... nevertheless, I continued to monitor their status during the game... the AI was using hunt commands... move commands... reverse commands... everything.. to try and get them out...

Finally, they all manuvered themselves into weird positions where they appeared to be frozen and couldn't move... but, if I had tried to cross that bridge... I am sure they would have shot me in the belly...

A small suggestion... perhaps Battlefront should examine this phenomena... and with a patch now... or a future change to the game engine... insert a routine or something such that if a functional tank does fall off a bridge... into water or solid ground or whatever... it will sustain some form of damage... like immobilization... gun damage... crew casualties... something like that...

Cordialement, Duke of Earl

[ January 08, 2003, 03:12 PM: Message edited by: Duke of Earl ]

[Bruno Weiss]

Probably not within the scope of CM. What happens after a tank falls off a bridge would take considerable time to ascertain, and therefore would be outside the 30 mins or so battle scenario.

[MikeyD]

Congratulations. You've managed to come up with one of those rare anomalous situations. For the most part, tanks don't get damaged when they fall off bridges because they don't fall off bridges! I'd be interested to hear if anyone else has ever managed to knock a live functioning tank off a bridge.

[SpineRipper]

I don't get your reply BW. How can falling off a bridge not fall with in the scope of a 30 min battle. I mean falling is a here and now event not something that does not take vary long to assertain what the result's were.

The bigger problehm for the above described event would be if thier are VF on the other side of the bridge. A few tanks fall off a bridge, with nothing happening to them, now when you try to cross the bridge they blast the crap out you. You might not even be able to fire back scine the evevation is greater than the depression of tank main gun's.

Well nothing ment by anything, just the thoughts of a mad man.

[karch]

Pretty sure gun elevation and depression aren't modeled, so you could shoot them even if they were just about beneath you from what I remember reading

[Bruno Weiss]

I guess you could still shoot them, but what happens if they land on their top, or landed with the barrel stuck in the mud with the hull dangling in the air? Anyone know the relative armor thinkness of the bottom of a tank? Is that in the database?

[Spook]

Originally posted by Bruno Weiss:

Probably not within the scope of CM. What happens after a tank falls off a bridge would take considerable time to ascertain, and therefore would be outside the 30 mins or so battle scenario.

No, it is very much in the scope of a CM scenario, and doesn't take more than 30 minutes for the crew & vehicle to be affected in some way.

What is most likely the case here is that the CM engine can't yet assess the effect to a vehicle that makes a sudden vertical drop. Nor does it attempt to see if vehicles will "roll" like from going up a steep embankment or over a cliff. The latter case, IMO, would be more likely to occur than for vehicles going over a bridge.

Perhaps the future rewrite can assess "shock" and "immobilization" (or at least bog) to a vehicle that experiences a vertical drop or is assessed to "roll." Then again, for heavier tanks, I wouldn't mind too much the ability of such to MOVE into and through smaller buildings, with the risk of generating building rubble and vehicle immobilization.

PS: Greetings to Duke of Earl. I suspect from your sig that you are the self-same Duke at the Breakaway and GBRC forums.

[ January 08, 2003, 02:18 PM: Message edited by: Spook ]

[Duke of Earl]

Bonjour fellas,

Yep Spook... that be me...

THE PAW OF THE TIGER

SPOILER ALERT - SPOILER ALERT - SPOILER ALERT

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

Page 59 of the CMBB Manual says that bridges across small streams are 2 elevations in height...

Just like the bridges in this scenario...

Uh... correction... the T-34's were driving back and forth underneath the span in the gully... so maybe this bridge was 4 elevations in height... whatever... it was tall...

Page 79 of the CMBB Manual says that steep slopes are impassable terrain... so that means that the T-34's fell off the middle of the bridge and did not slide down the slope because the AI would not allow them to move that way...

Page 161 of the CMBB Manual says that a single elevation level can be 1.25 or 2.5 or 5.0 meters in height... depending upon the parameters of the map...

And it would appear to me that the drop from the main bridge in this scenario was at least 5.0 meters... that's over 15 feet ya'll...

So, I would think that the effects of a 5.0 meter vertical drop off a bridge... on a multi-ton T-34... would be rather instantaneous...

If I jumped off the same bridge on to a frozen stream you can bet it would cause an immediate effect on me... ouch!

And if there wasn't a downright flip over of the tank... then there should be some form of severe shock damage to the crew, the tracks, the gun optics, the gun mechanism, the hull, or something...

Fair is fair.. makes sense to me... considering how quick the AI is to bog down all of my tanks on dry land and immobilize them...

Cordialement, Duke of Earl

[ January 08, 2003, 02:52 PM: Message edited by: Duke of Earl ]

[SgtMuhammed]

It shouldn't be too hard to enable this in the engine. It could be something as simple as if a vehicle looses more than X number of meters in less than Y meters of movement then it is judged to have fallen and should have about a 90 percent chance of an immediate immobilization and crew shock as well as a good chance for a gun hit. I won't say it should be 100 percent because I have seen tanks take nasty spills and then drive away once the crew decided that they were still alive. :eek:

I think this would add an extra strategic option to the game. You know, something like hmmm.. should I risk trying to bust something through that parking lot on the bridge or do I risk falling off or what about that narrow ledge can I sneak around there or is it too dangerous. I would also love to be able to blow bridges but that is for a different thread . All in all CM does an outstanding job at modeling almost all the normal events we expect to see on a battlefield (if you can call a battlefield normal ) but there are some unexpected things which can have a fairly large impact.

Another OPFOR story:

While we were fighting the Dutch they had borrowed some of the Bundeswehr's Leopard II A6 tanks. During the rotation one was rolling along a narrow path on the top of a retaining wall for a retention pond (the Germans build them for flood control). The rotation was during the winter and it was pretty muddy but the pond was empty. The loader was in the TC hatch because the TC was in the loaders compartment to write an Op-Order (because there is more room on that side of the tank). Well either the driver wasn't paying attention or the path got to narrow, regardless the tank slid down the wall and hit on its side throwing the loader out of the hatch and into the pond. The tank then proceded to turn turtle with the turret landing squarely on the loader's legs. Amazingly though the mud on the bottom of the pond was so thick that the tank just pushed his legs about a foot into the mud. He got some bruising but other than that was perfectly fine (physically at least ). Everyone inside the tank was bruised up but ok as well.

Ah the joys of army life.

[ January 08, 2003, 02:44 PM: Message edited by: Sgtgoody ]

[Guest Sgt. Emren]

I'd be interested to hear if anyone else has ever managed to knock a live functioning tank off a bridge.

Well... I haven't personally done so, but while stationed in Bosnia in '96 with NATO, a Leopard tank skidded off an icy bridge and landed upside-down in a small river. Apparently (and I'm no tank expert), because of the way the turret was angled when it happened, the tank commander was trapped in the turret and could not get out of the bottom hatch. The rest of the crew where relatively unharmed, but the tank commander suffered a grim fate. He drowned in the ice cold water. He was Corporal H.S. Jacobsen, 28 years old.

So it CAN happen. The tank, of course, was recovered, so it wasn't exactly destroyed, but it was certainly knocked out temporarily.

[Bruno Weiss]

I think the rationale would probably go something like; "well, tanks are not supposed to fall off bridges, therefore besides the fact that they can, to model the end result would be outside the scope of CM, because their not supposed to fall off bridges in the first place." Therefore, for those with high hopes of seeing something akin to Gomez' train wrecks, the most likely thing that would happen, if anything at all happens, would be that the first rule of the most simplistic fix would apply. Ergo, BTS would simply fix it so that tanks do not fall of bridges. Ya think.

[Dan Robertson]

Actually a drop of 5m is sustainible depending on how it happens.

In fact one of the first tanks was demonstrated "jumping a tall building in one bound" to King Geroge V in WWI. The tank went over a reproduction of a "German bunker" which was about as high as the the tank was long ~8metres.

The tank was undamaged but all but one of the crew was knocked unconcious in the fall.

Incedentally part of the tests on Challenger II the british armys new tank included driving off a vertical drop of about 5metres or so without damage. It would depend on how you tok the drop and the factor would be uncomputable. Hence it isn't in the game.

For example a t34 drops of bridge at and angle of 21 degrees fall 4 metres, lands on its side looses a track, the crews heads all hits somthing hard. For the purposes of a battle it is knocked out.

Or T34 tank falls off bridge at angle 21 degrees fall 4 metres. Tank rolls full circle onto its tracks, crews head don't hit anything hard. Tank carries on after a short delay.

[laxx]

aha.... the possibilities are endless for Gamey tactics..fall off the bridge and attack!

[Duke of Earl]

Bonjour fellas,

In fact one of the first tanks was demonstrated "jumping a tall building in one bound" to King Geroge V in WWI. The tank went over a reproduction of a "German bunker" which was about as high as the the tank was long ~8metres.

You know... I think I saw that film clip on the History Channel... it was like a Mark IV Female I think... and it flipped end over end before stopping... those guys should have worn seat belts...

I still find it hard to believe that a CMBB tank of any description... let alone a T-34... can fall off a bridge onto a frozen stream and not have any damage to any of the systems or the crew...

I think I'll start using them in a 'bungee cord' manuever... I'll deliberately drive them off bridges to avoid counterfire... and then drive them up the gully or river bank to attack from the flank... this is certainly a real tactical possibility with the present game architecture...

Cordialement, Duke of Earl

[Spook]

Originally posted by Bruno Weiss:

I think the rationale would probably go something like; "well, tanks are not supposed to fall off bridges, therefore besides the fact that they can, to model the end result would be outside the scope of CM, because their not supposed to fall off bridges in the first place." Therefore, for those with high hopes of seeing something akin to Gomez' train wrecks, the most likely thing that would happen, if anything at all happens, would be that the first rule of the most simplistic fix would apply. Ergo, BTS would simply fix it so that tanks do not fall of bridges. Ya think.

But what if a tank is KO'ed on the bridge? Wouldn't one want the option to have it pushed off by a comparable or larger vehicle with sufficient power?

Granted, I don't expect for this issue to be addressed in CMBB. But any assertion that vehicles "falling" or "rolling" from bridges or other sudden height drops (like over a cliff) is beyond the scope of a future CM engine rewrite is premature at this time.

In principle, it MIGHT be a straightforward matter to address; vehicles dropping or rolling would check for crew shock and vehicle bog/immobilization. After all, as it stands in CM now, certain conditions can exist as that a vehicle can bog/immobilize in ANY terrain save a paved road, thus being relatively "pervasive" in application. The event of a drop/roll IN THEORY seems as something that could be added in on the calculation side.

Of course, some would also want to see the visual model too of a CM vehicle flipped over or sumfink after a drop/roll. From my end, this visual representation wouldn't be necessary.

[ January 09, 2003, 12:49 AM: Message edited by: Spook ]

[MrSpkr]

SPOILERS

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

*

I am the designer of that scenario. Let me see if I can shed a little insight.

The bridges are 4 levels in height. This was necessary (though not completely desireable) for map asthetics.

The troubles you are seeing with the tanks falling off the bridge(s) is one of the primary reasons that scenario is recommended for play as Human Soviets vs. Axis AI. During playtesting, it was noted that the AI pathfinding, combined with the variety of Soviet tank crew ratings (veteran, green, conscript, etc.) placed the AI at a severe disadvantage in properly developing the Soviet attack. The AI seems completely inable to coordinate the large number of variable speed tanks that all have to pass through a narrow corridor. Numerous attempts to rectify this all failed or were deemed unsuitable (for example, placing the Soviet tanks on the German side of the gully was considered, but rejected as unfair to the Germans).

Also, I considered making all the tanks the same experience level to achieve some speed uniformity. This did not work for two reasons. First, aesthetically it was simply not as pleasing. Second, and more importantly, the variety of Soviet tanks involved in the scenario had different speeds simply because they were different kinds of tanks -- so making the experience ratings identical would not solve the problem.

As a result, the AI clogs up the bridge with too many tanks, resulting in some being pushed over the side by other tanks trying to move through. It becomes a real problem.

I suggest that, if you play it solo, you try it from the Soviet side. It is extremely challenging, and will give you some idea of just what the Soviets faced when the Tigers first came out.

Steve "MrSpkr" Hines

[ January 09, 2003, 01:38 AM: Message edited by: MrSpkr ]

[SgtMuhammed]

Sgt. Emren

We also had a not so happy event here. An M1 during winter night ops drove into a large water filled hole and got stuck. Water began filling up the driver's compartment but because of the way the turret was turned they couldn't get the man out through the turret and they couldn't get the tank out of the hole for almost two days. Sorry I don't know the soldier's name though, they didn't release it.

Back to the thread...

While most of us know that there are a lot of variables to tanks falling, rolling, sliding, what have you, down cliffs and off bridges, I think most of us would be happy with just a general, if you fall off something in an AFV you will be out for the fight type rule. While it may not happen often the fact that it has happened and can be recreated without too much difficulty shows that this is a valid issue. I think going the other way and not allowing things to fall at all would be even more undesirable than now.

[Bruno Weiss]

Well, there is something to be said about bridges being blocked by wrecks, but okay then what about the problem of "how" exactly a vehicle (tank), falls off a bridge. Would it always fall off like a cat and land on its treads? Or is the AI supposed to calculate the gravitational and geometric forces involved that cause an object to rotate in mid-air while falling to occur, and then subsequently determine at what precise angle the object will then hit the ground? Assuming it lands like a wet buffalo chip with a splat and doesn't roll, which to simulate such a roll would take even more AI calculations.

Big difference for the AI processing. And when is the last time anyone has seen the AI allow a vehicle to flip over?

Always landing on the treads and being knocked out would I should think, be only slightly less gamey than it always landing on its treads and being okay. Even that much would probably require some hard coding. The insurmountable BTS block to kiddome fantasys. (You'll shoot yer eye out kid!).

[ January 09, 2003, 11:02 AM: Message edited by: Bruno Weiss ]

[Spook]

Originally posted by Bruno Weiss:

Well, there is something to be said about bridges being blocked by wrecks, but okay then what about the problem of "how" exactly a vehicle (tank), falls off a bridge. Would it always fall off like a cat and land on its treads? Or is the AI supposed to calculate the gravitational and geometric forces involved that cause an object to rotate in mid-air while falling to occur, and then subsequently determine at what precise angle the object will then hit the ground? Assuming it lands like a wet buffalo chip with a splat and doesn't roll, which to simulate such a roll would take even more AI calculations.

Big difference for the AI processing. And when is the last time anyone has seen the AI allow a vehicle to flip over?

Always landing on the treads and being knocked out would I should think, be only slightly less gamey than it always landing on its treads and being okay. Even that much would probably require some hard coding. The insurmountable BTS block to kiddome fantasys. (You'll shoot yer eye out kid!).

All fair questions, and to what level of abstraction is acceptable will not be resolved at this point in time, even IF Battlefront does want to pursue the matter.

And yes, I expect that accounting for this (given that I am not constraining to bridge events but also cliffs, slopes, and ditches) can be addressed by hard-coding. Thus why I wouldn't expect a POSSIBLE solution to the issue until the future rewrite.

So you watched a certain "Christmas" movie too, huh?

[Spook]

Added note.....

Originally posted by Bruno Weiss:

Big difference for the AI processing. And when is the last time anyone has seen the AI allow a vehicle to flip over?

Both the AI and I do it all the time in GTA III. GREAT fun to play last year.

[Duke of Earl]

Bonjour fellas,

The Paw Of The Tiger is a very good scenario... quality of this scenario has never been the issue...

And yes, the AI should be chosen to play the Axis... just as it says in the scenario notes... but I can't resist Tigers... anyway, as we all know, the AI is substantially better on defense than offense...

Based upon previous comments... and in agreement with same... I would suggest a very simple solution to this phenomena... if such a solution is possible... like this:

A. Tank model falls off bridge (or other high object)...

B. Tank model lands on tracks (upside down tank not required)...

C. Tank model suffers various degrees of damage to its systems and crew (depending upon the height of the fall)

Nothing complicated is required... remember K.I.S.S... so, a simple abstraction will do... like rolling dice and using the tables in Squad Leader...

Now it is true that... in the real world... a multi-ton tank MIGHT escape a 5.0 meter fall with little or no damage to vehicle or crew...

But, as confirmed by the designer... the bridges in this scenario were 4 elevations high... which is probably 10.0 meters in height... and that's over 30 feet ya'll... so, you do the math...

Cordialement, Duke of Earl

P.S.- It just occurred to me that if a tank can fall off of a 10 meter bridge... then perhaps a squad can also fall off a 10 meter bridge... I shall have to check this out...

[ January 09, 2003, 12:20 PM: Message edited by: Duke of Earl ]

[Bruno Weiss]

WHAT I WANT FOR CHRISTMAS

I want a tank what falls off bridges and such, turns over and rolls down hills at least 15 feet and sustains absolutely no damage what so ever.

F F F F F F F F F F F F F F F F F F F F F F F F

You'll shoot yer eye out, you'll shoot yer eye out...

[Dr. Rosenrosen]

I think this bug/phenomenon is caused by the fact that the map is not truly 3-D, i.e. the engine can't model items both on and below a bridge.

Just today I discovered a similar situation that results from this problem. I started a scenario with a frozen river and tall bridges. I wanted my ski troops to follow the river, but wasn't sure how they would react to crossing under the bridges. I sent some in Run mode and some in Move. All of the units immediately jumped up on the top of the bridge when they crossed it, even though it was 4 height units up. The ski troops that were running immediately became exhausted as soon as they touched the bridge. (I guess it's tough jumping up 10m or more and cross country skiing on pavement. ) The troops that were moving fared much better. None of them lost their skis, and they all return to river level once they were across the bridge.

It seems that the engine doesn't have a logical way to handle units passing through a bridge tile from any direction except the "correct" one for crossing the bridge. Given the nature of this bug, I expect we'll have to wait for a rewrite instead of just a patch.

Dr. Rosenrosen

[murpes]

Originally posted by Bruno Weiss:

Big difference for the AI processing. And when is the last time anyone has seen the AI allow a vehicle to flip over?

And here is one of the major problems with reading the forums - limitations of the game engine are pointed out that I never noticed before! It never occured to me that all vehicles were upright all the time - but now I'll be aware of it. Ignorance is bliss sometimes

[Bruno Weiss]

Precisely! And then if tanks could turn over, we'd have the problem of the upsidedown hull rotating towards the target while the turret was stuck in the mud. See what happens when you start thinking outta the box.