Unbelievable!!!

[Holien]

This is problem because it denies otherwise valid tactics like using at-mines to block narrow corridors.

The AT mines in themselves would do that?

[Broadsword56]

With the upcoming Market Garden module the "ghosting" will be more annoying with some scenarios. The only thing that I can think of that might work is to impose a time delay on the first vehicle that "ghosts" thru a destroyed vehicle. Maybe three or four turns or so. This would kinda simulate moving the destroyed vehicle. It would normaly take much more time to move a destroyed tank but at least there would be a penalty.

I'd want to hear more specifics of the situation you're imagining, but I'm thinking about the raised portions of Hell's Highway or whatever road is used as "Club Route" into Arnhem...

Four minutes for someone in authority to see the problem, issue the order for a blazing wrecked vehicle to be pushed off the road, under enemy fire, with its ammo likely cooking off all the while? I'd be skeptical of that. (If only AAA could respond that well in the peacetime 21st Century!)

Maybe pushing a still-flaming wreck off the road with a regular tank might have been considered too dangerous anyway -- XXX Corps had dozer tanks for purposes like that. But, given Murphy's Law, that dozer tank is likely never to be in the place it needs to be when a wreck happens, and probably has to fight its way through a congested column of traffic to reach the wreck site.

All of which is my way of saying wreck recovery/pushoffs/dozer tanks are something properly beyond the time and scale of CMBN. Do whatever you like in your games, but in mine I expect I'll be using a "wrecks block roads and single-lane bridges for the duration of the battle" house rule for Market-Garden (depending on the specific terrain and situation of course.)

[slysniper]

Who made the grand decision that tanks would not push or tow other tanks out of their way, saying the game is wrong for letting units move past wrecked units blocking the path.

Or that tankers would patiently wait for a dozer tank to come along and clean up the mess.

Lets put you in a platoon of tanks on a one lane road with no exit other than on the road forward or back.

A enemy AT gun takes out the lead tank, then the second before your remaining three tanks destroy it. No other fire seems to be coming in and your mission is to lead and clear the roadway to the next junction. what are you going to do.

Well if its me, I dont want to sit there very long, figuring if the enemy had a AT gun focused on this point, likely Arty will be raining down soon or they might be pulling up reserves to finish the job that has been started.

I figure most would either use their tanks to try and push the units out of the way enough to progress or tow or something. Better chance that than any other option. If things go bad and you wreck your tank, oh well, still a better option than what others are saying here.

So maybe all the game needs is a feature that if you are moving through wrecked equipment, there is risk added for immobilization like is already present in the game for terrain.

Say , if a tank passes through a truck , not much risk. But if it has to pass through another tank, a very high risk, like a 30% it also will be immobilized.

That would make a road jamb real quick. Anyway, you get the concept.

[Georgie]

I'd want to hear more specifics of the situation you're imagining, but I'm thinking about the raised portions of Hell's Highway or whatever road is used as "Club Route" into Arnhem...

Four minutes for someone in authority to see the problem, issue the order for a blazing wrecked vehicle to be pushed off the road, under enemy fire, with its ammo likely cooking off all the while? I'd be skeptical of that. (If only AAA could respond that well in the peacetime 21st Century!)

Maybe pushing a still-flaming wreck off the road with a regular tank might have been considered too dangerous anyway -- XXX Corps had dozer tanks for purposes like that. But, given Murphy's Law, that dozer tank is likely never to be in the place it needs to be when a wreck happens, and probably has to fight its way through a congested column of traffic to reach the wreck site.

All of which is my way of saying wreck recovery/pushoffs/dozer tanks are something properly beyond the time and scale of CMBN. Do whatever you like in your games, but in mine I expect I'll be using a "wrecks block roads and single-lane bridges for the duration of the battle" house rule for Market-Garden (depending on the specific terrain and situation of course.)

I was considering any situation where a destroyed AFV would block the only road forward. I agree, it would probably take more time than the length of the scenario to move most destroyed AFVs. But, a delay of some sort would be better than no delay at all and it would still be good tactics to take out the first tank in the line but it would not effectively end the battle. Of course those that wanted to could agree on house rules for their games.

[George MC]

Who made the grand decision that tanks would not push or tow other tanks out of their way, saying the game is wrong for letting units move past wrecked units blocking the path.

I'll take a real stab in the dark here and say - the developers?

[George MC]

On a more serious note it was discussed and on balance game play issues i.e. one single KOd tank blocking a crucial road/bridge etc could screw up the whole game. Given at the time this was a totally new experience I think the general consensus us was as you see now. Possibly as the game and players mature it might be a different context. Still whatever decision is made it's BFC's - their game and ultimately their choices. Simples.

[YankeeDog]

All of which is my way of saying wreck recovery/pushoffs/dozer tanks are something properly beyond the time and scale of CMBN.

Eeehhh... not necessarily. I've read multiple accounts of regular tanks (not dozer or other engineering vehicles) pushing friendly disabled tanks out of the the way. Bear in mind, I'm talking about nudging the vehicle a few feet onto the verge so others can pass, not towing it all the way about to the depot (though I actually have actually read of regular tanks doing tow service in a pinch, too).

Then again, I've also read accounts of entire columns being held up for quite a long time by a single disabled tank at the front. So that happened sometimes, too.

Devil is obviously in the details -- Is the hulk burning? what are the road/ground conditions like? How far does the hulk need to be moved to clear the road? Is the running gear of the wreck still more or less intact? Etc.

Overall, I'm not sure it's any more realistic to assume that a burning tank hulk on a road would *always* block the road for duration of a CM-scale battle, than it is to assume it the hulk would be no significant obstacle on the CM-scale.

I Dunno. This seems like one of those tricky areas where where there is no easy solution; I'm not sure there is any easy, unambiguous "house rule" is that is going to create a more realistic situation than the game engine does currently stands.

[Broadsword56]

I see your points

And when it comes to game setups, whatever floats your boat.

But, given what we now know about this issue, I'd rather have a house rule for my own battles to fill in what I might see as a realism gap in a particular situation, than to trash the game and developers or demand they do the impossible to "fix" it.

[cool breeze]

Ok I know this idea is a little silly, and there are better ideas related to this game I could mention instead, but I just got this idea so here it goes.

I understand that getting a vehicle out of the way would involve another vehicle. But I imagine it would usually also involve squishies doing important things like attaching chains, digging, putting down planks ext. I think it might be a good quick and dirty compromise to make it so engineers could disappear a destroyed or empty vehicle, given enough time. It would work like blast or mark mine but take way longer. I know it would be weird to have some infantry able to "move" a tank, but it seems to me a better compromise than having tanks do it. for one thing it seems like to do it in the real deal you'd need both, squishies and a vehicle to move it, but as the infantry are softer, and since a friendly vehicle is probably near by ( which is why you want to get rid of the blockage) its seems more fair to make the inf. do it.

[YankeeDog]

I see your points

And when it comes to game setups, whatever floats your boat.

But, given what we now know about this issue, I'd rather have a house rule for my own battles to fill in what I might see as a realism gap in a particular situation, than to trash the game and developers or demand they do the impossible to "fix" it.

As long as you're playing with rational people with whom you feel comfortable negotiating the grey areas of such house rule, I think this is a reasonable point of view.

[YankeeDog]

I understand that getting a vehicle out of the way would involve another vehicle. But I imagine it would usually also involve squishies doing important things like attaching chains, digging, putting down planks ext. I think it might be a good quick and dirty compromise to make it so engineers could disappear a destroyed or empty vehicle, given enough time. It would work like blast or mark mine but take way longer. I know it would be weird to have some infantry able to "move" a tank, but it seems to me a better compromise than having tanks do it. for one thing it seems like to do it in the real deal you'd need both, squishies and a vehicle to move it, but as the infantry are softer, and since a friendly vehicle is probably near by ( which is why you want to get rid of the blockage) its seems more fair to make the inf. do it.

I'm not sure if this is a good assumption to make.

First of all, this kind of thing actually not typically a combat engineer task, but rather a vehicle maintenance/recovery team task -- such teams were on the TOE of most armored units at the battalion or brigade/regiment level. Combat engineers could probably serve the turn in a pinch, but the tank crews themselves would probably know as much or more about how to do this sort of thing as engineers would, and the tools for basic movement & recovery (tow cables, etc.) were usually carried on the tanks. So if you're going to let combat engineers do this, you should also probably let a nearby tank crew dismount and execute the recovery.

If the vehicle were still at least somewhat able to roll on its own running gear, another vehicle of similar weight and power could probably nudge it a few feet quite easily, without setting/attaching planks, tow cables, or whatever. This would probably be the case for many vehicles hit on the hull and/or turret.

On the other hand, a burning hulk with an entire track and much of the running gear blown off of one side might be very difficult to move indeed, requiring substantial prep work and the use of a specialized recovery vehicle with a power winch and/or crane. Soft ground would make any recovery task much more difficult.

[c3k]

I want destroyed vehicles to rust into nothingness. Say, about 4 minutes for a tank to oxidize back into elemental iron-ores, and 1 minute for jeep. That simulates the push-away time and lets the "ghost-through" occur with no suspension of disbelief.

Yeah, rust catalyst. THAT's the ticket...

^^^

All tongue-in-cheek. Although that is a "solution", it's not one that this game milieu would support. This is a near-simulation; therefore, when the physics are not near-realistic, there are complaints.

There are solutions: some may not be acceptable. Some are worse than others. You may not accept the current solution, but it is far better than some.

[JonS]

a delay of some sort would be better than no delay at all

I'm fairly sure there is a delay currently, in that vehicles "ghosting" move much slower than they otherwise would over given terrain.

[Holien]

I'm fairly sure there is a delay currently, in that vehicles "ghosting" move much slower than they otherwise would over given terrain.

FYI I did not get the Panther through the gap and there were certainly delay issues. But without testing I can not say this would be case all the time.

IMO IRL the Panther would have nudged the half track aside as it was not 90% to gap and a little sideways shove would have been possible.

In the Game the Panther rotated 360% and ended with it's rear facing the gap. Which if there had been someone behind it would have been terminal!! As it was the so**ing Zooka Teams were dead or running for their lives.

BTW this is all your fault Jon! Your scenario!!! I won't reveal name to avoid spoilers for people.

[JonS]

Serves you right for being a gamey edge-hugger

[Holien]

Serves you right for being a gamey edge-hugger

Well if designers don't make big enough maps what do you expect!!! All I was doing is trying to dodge minefields.. (Which BTW I failed!!!)

[BLSTK]

Serves you right for being a gamey edge-hugger

What you need is a bigger computer monitor. Problem solved.

[agusto]

agusto, just in case you miss it, I replied to your ArmA 2 post a few posts up.

Yes, thanks for the reply. I didnt miss it, just wasnt online. I am personally not botherd by the clipping, i always just wondered why it was made that way.

So, can we conclude from what you have said in this thread by now that the clipping issue is the result of inadequate processing power available on todays computers to handle pathfinding on CMs scale and complexity? If yes, that is good, because then it can only be a question of time until the problem can be solved, isnt it?

From my experience - the tac AI both friendly and enemy - appears to be absolutely non existent in Arma - the primary reason I stopped playing it. I played a scenario with the british forces module and when ambushed in a Scripted mission - the 8 AI guys with me stood around and did nothing. Not very little. Absolutely nothing. When I compare that to what the Tac AI does in CM - the two are incomparable.

[Battlefront.com]

Good discussion, though we've had this one more than once. I think Phil and others have adequately explained the differences between how we would all (including me) like to have CM play and how unlikely that is to happen due to technical limitations of today's computers. Not limitations of what our programming team can do, as evidenced by the lack of anybody dealing with the twin issues of clipping and path blocking any better than we do (depending on perspectives, most are a lot worse).

The clipping problem is unavoidable because 1:1 polygonal collision detection for this scale of combat is simply beyond what computers can handle given CM's level of graphical detail, number of units moving at any one time, and environmental diversity. Because it is physically impossible to get the computer to handle it the way we want, might as well forget arguing for change that will never come.

Instead, all we can do is detect collisions in a more abstract way, then assess some sort of penalty to at least approximate what would be happening in real life. The major form this takes is slowing down movement. In your heads you can think of this as either carefully driving around a tight obstacle or nudging it a bit or doing both. Either way it's causing less than ideal behavior compared to if the blockage wasn't there.

We could allow a blockage to remain in place and absolutely (instead of just by chance) completely stop movement through that location. In fact, long ago CMx2 was coded this way. It made the game, at times, unplayable. Whatever frustrations are expressed here, I can promise you there would be a lot more unhappy people that found themselves stopped cold from any chance of advancing because they lost their lead tank in a bad spot.

OK, so then why not simulate moving dead vehicles? Wouldn't that be more realistic? If we did a detailed simulation of it, sure it would be. But anything short of that would be no real improvement over the way things are now.

In real life the circumstances for moving a knocked out vehicle are extremely complex and situationally dependent. In another thread someone just lost 26 guys to a Sherman that catastrophically exploded. I was reminded of my own early experience of this back in the CMSF Beta days when I was moving some troops near a burning T-62 that suddenly had a secondary explosion. In short, getting near enough to a cooking vehicle to move it is dangerous. Getting out and hooking up tow cables is even more dangerous. Yet tow cables (or tow bars) are often exactly what is needed to move a vehicle.

When a vehicle gets knocked out it's not as if you can be sure it will roll when pushed. If the transmission was in gear, or got damaged, then the vehicle won't roll at all. Pushing a vehicle that doesn't want to roll may be successful, but it probably won't be unless the pushing vehicle is substantially larger. It also needs sufficient traction, which often isn't available. And this is assuming the vehicle is in a position that is optimal for pushing AND where it winds up won't further complicate things. Very frequently pushing is actually makes things worse and instead you have to pull/tow something, which gets into another area of complication.

In short... by and large when a vehicle of any significant size gets knocked out it stays there until recovery and/or engineers move it. And that is the sort of thing that isn't likely to happen within a CM battle's timeframe. In fact, it generally would require a separate action completely, such as expanding a "bridgehead" of infantry around the area so the blockage could be safely cleared.

And let's not forget that we would also have to create the opposing player AI and TacAI sufficient enough to deal with all these factors. Oh that's a bunch of fun

All in all I think CM is not that out of line with reality as some may think. Graphically? Yes, but if that's a "deal breaker" then you're not going to have much to play with for the next 5-10 years. Gameplay wise? We're probably too liberal in some situations, but overall that's the way gamers want it to be. "Fixing" it for some would "break it" for a much larger crowd. And to do it "realistically" involves a lot more diversion from other improvements than I think anybody would be interested in suffering through. It's a major effort, hence why it is the way it i snow

So do not expect much to change

Steve

[James Crowley]

O.K. That seems to cap it off effectively, which is a little disappointing.

Having started this thread, I would like to take the time, firstly, to put a few things straight.

There was no anger, angst or anything else of a similar nature. The term ‘bollocks’, in this context, has the same meaning as ‘crap.’ While that may seem, by some, as being an overly harsh criticism of the gap/bridge/ghosting phenomena, it stemmed from passion rather than anger.

I have already said that my comment, that it was a game-breaker, was overstated. I have supported and enjoyed the CM series since its inception and have bought every CM-related product other than the SF modules, as I have no interest in Modern, and will continue to buy the Gustav Line, East Front etc. I am a huge fan of CM and think that it has come on in leaps and bounds from its CMx1 days.

However, I do think that this, shall we call it problem, detracts from what is a very detailed and accurate simulation/game. IMO It also goes a long way to negating the concept of choke-points, which seem fairly central to military theory and practice. And, unless my less than perfect memory is playing tricks, it didn’t happen in CMx1.

While I can readily accept that, in some instances such as the current illustrated game, this could be interpreted as a bit of pushing, pulling, squeezing and so on, that explanation does not hold water with a vehicle blocking a single track bridge or road with no possible passing point. As a result, it is quite possible that real-life tactics will be over-ridden, as it will never be possible to seal-off a choke point; vehicles will just be able to ghost through.

I have just set up test scenario with a single span bridge. With a single tank parked on the bridge both of the follow-up tanks ghosted through with very little evident delay. It is really difficult to square away this situation. It is not approximating squeezing past because that is impossible nor, apparently, is it pushing; so how is it to be interpreted? In short, as far as I can see, it can’t and that has a negative impact on the reality of the simulation and ones immersion in it.

I have no technical knowledge in the area of programming, so will happily defer to those who do.

However, might it be possible to utilise, to some extent, the existing collision routines?

Currently, when vehicle paths cross each other, either because of mismanagement or reaction to enemies, the vehicles usual stop, temporarily, and then either continue on or change course. Often that course change is less than optimum in terms of what the player wants and leads to frustration.

Would it be possible, when potential collision is detected, to have the vehicle stop dead? Fully, no further movement. That would have a number of possible effects.

Firstly, I think but stand to be corrected; while an imminent potential collision is detected there would be no further movement. So that might stop vehicles ghosting through?

Secondly, it would stop vehicles heading off in, sometimes, random directions; to the consternation of the owner.

Thirdly, it would have the potential to slow vehicle movement down considerably, especially in crowded or congested situations. Personally, I would have no problems with that but accept that others may.

Just some thoughts and I’m sure that either Phil or Steve will happy to correct me.

[Vanir Ausf B]

And, unless my less than perfect memory is playing tricks, it didn’t happen in CMx1.

It did, but differently. CMx1 did have collision detection and vehicles could push each other. I am puzzled as to why it is now thought to be too difficult to do.

The catch is that any vehicle could push any other vehicle, i.e. a Kublewagon could push a King Tiger off the road just as easily as the other way around, so in practice the result was not much different than what we have now in CMBN.

[Erwin]

+1. The effect is the same. It's just that the graphics of this are not optimal. Not a biggy. There are more important items to address.

[Battlefront.com]

O.K. That seems to cap it off effectively, which is a little disappointing.

Nobody is more disappointed than us in the limitations of technology (in this case, mostly that) and/or development resources. Not only would we love to do things like this to make the game better, it really stinks to be held accountable for reality breaking conditions which are outside of our control. And I don't just mean outside of Battlefront's control, but outside the control of games with $50,000,000 budgets. We can't do cold fusion any more than some of the things you guys think we should be able to do.

There was no anger, angst or anything else of a similar nature. The term ‘bollocks’, in this context, has the same meaning as ‘crap.’ While that may seem, by some, as being an overly harsh criticism of the gap/bridge/ghosting phenomena, it stemmed from passion rather than anger.

I for one got that first read through, so no problemo.

However, I do think that this, shall we call it problem, detracts from what is a very detailed and accurate simulation/game.

No argument there. But there's a long list of "detractors" when you compare CM to reality. Especially the older CM games since they were far more abstract and graphically primitive. At some point we just have to accept the limitations and don't allow them to get in the way of enjoyment. I can list far more "detractors" in Gary Grisby's Kampfgruppe on my old Atari 800 than I can for CMBN, yet I enjoyed the heck out of that one.

IMO It also goes a long way to negating the concept of choke-points, which seem fairly central to military theory and practice. And, unless my less than perfect memory is playing tricks, it didn’t happen in CMx1.

Playing tricks I'll explain that below.

Would it be possible, when potential collision is detected, to have the vehicle stop dead? Fully, no further movement.

Absolutely possible, though depending on the situation it could be extremely difficult for the game to handle. Especially in RealTime, which like it or not is an extremely important consideration since the entire game engine is based around realtime events. WeGo gets off a little easier, but it can still be ugly.

The problem comes down to path finding. In order for the game to function at its most basic level, there is a collection of path "maps" which the game uses for all movement functions. These are incredibly expensive to determine on the fly so a master list of routes, so to speak, is established ahead of time. There are different master maps for different classes of units, too, since what works for infantry doesn't necessarily work for a heavy tracked vehicle which might not work for a light wheeled vehicle, etc. Change ANY one variable, such as blocking a bridge with a light vehicle, and all of the pathing maps need to be updated. Even with absolute "this is blocked for everybody" it's messy, but to have conditional blockages it is really bad.

I'm not a programmer so I can't comment on why this or that condition is more or less difficult for pathing to be altered. What I do know is that temporary, conditional blockages present unique challenges which something like a boccage breach do not.

One thing I do know of is compounding cause/effect issues. Adding an additional access point (breach) means modifying the path maps to be more flexible. As with most anything, it's generally easier to add capabilities than to remove them. Which is to say it's a lot easier to say "oh, vehicles can now do this in addition to that" instead of "now vehicles can not get to where they once could, and that's not acceptable so I need to try as hard as I can to find an alternative. Further, there is a practical need to find an alternative that doesn't totally suck". The only way to find these, if they even exist at all, is to crunch numbers and chew up lots of RAM. A lot of numbers and a lot of RAM.

Think about driving in a city when it's construction season. Of if you're from Boston, when the Big Dig was at its worst. Far, far easier to have a street reopened than to have one closed off. Similarly, there are fewer complaints about driving in cities at 3 in the morning compared to 3 in the afternoon. Restrictions cause stress, be it to a driver's mental state of a computer's need to crunch numbers and chew up RAM.

That would have a number of possible effects.

Firstly, I think but stand to be corrected; while an imminent potential collision is detected there would be no further movement. So that might stop vehicles ghosting through?

Sure, but then we get into the problems I already spoke of. Namely the "be careful about what you ask for because you might get it". We had this in place back in CMSF and it sucked. Nobody liked it. And that was from people, like you, who value realism. Simply put it harmed the game, as a game, too much. Nobody wants to get invested into a game and 4 minutes into play have it physically impossible to either win (if on the attack) or be challenged (if on the defense). Which is why blocking stuff without any sort of hope of unblocking is not an option we will consider. We already did it and we know, for sure, it isn't what people want.

Secondly, it would stop vehicles heading off in, sometimes, random directions; to the consternation of the owner.

This is a result of a compromise between the vehicle stopping dead and then trying to figure out what to do (generally not a good idea) and having them "guess" at some direction with the presumption that any action is better than inaction. Obviously that's not always the case and this sort of behavior could certainly use some improvement. But having vehicles permanently stop isn't a viable solution.

Thirdly, it would have the potential to slow vehicle movement down considerably, especially in crowded or congested situations. Personally, I would have no problems with that but accept that others may.

Oh no it wouldn't What you'd get is what CMx1 suffered from quite badly in that it's impossible to control vehicles in any sort of situation where there's a unit anywhere near it. Think back to your CMx1 days of trying to do column movement on a road. Unrealistic, frustrating chaos was often the case. CMx2 isn't perfect either, but it's significantly better.

The more variables you throw into the mix the worse it becomes.

Back to CMx1...

It was impossible to block something in CMx1. Everything could be pushed. It is akin to "ghosting" in CMx2, but without the direct visual representation. Which is to say that, effectively speaking, there is no difference between the behavior in CMx1 and that of CMx2. At least not inherently.

The difference was in CMx1 you could fully clear a path and then eliminate delays to following units. This is not possible in CMx2 and it's by design since clearing a disabled heavy AFV from blocking a bridge is physically impractical within the scope of a CM battle. Overall I'd say that CMx2's system is more realistic, though perhaps not as satisfying on the surface.

Should "ghosting" carry more delay penalties? Perhaps. It's something that can be looked at in the future. It's about the only thing I can think of that has potential to be reasonably improved.

Just some thoughts and I’m sure that either Phil or Steve will happy to correct me.

If you think there is some clever way to get (at least mostly) the behavior you picture in your head, without any of the downsides you're not imagining, then you do in fact need to be corrected. Because it's fantasy and therefore unimplementable. If you understand that there's no way to get what you're imagining, and are instead willing to see an extremely small improvement over time (relative to the desired behavior), then that is likely possible.

Steve

[Battlefront.com]

It did, but differently. CMx1 did have collision detection and vehicles could push each other. I am puzzled as to why it is now thought to be too difficult to do.

It's not. We could do it but, overall, we feel it is less realistic and therefore not worth bothering with. What's difficult to do is to get significantly more realistic behavior. True there is way more computing resources now than when we coded CMx1. However, CMx1's level of internal detail was scaled to the hardware of the day as is CMx2's.

One slight correction. CMx1 didn't have poligonal collision detection at all. CMx2 does have it, but it has to be "forgiven" in certain complex situations (for example two tanks moving into each other) because it's too computationally expensive. Same thing for the big budget FPS games. Theoretical collision detection capability doesn't mean much if it crushes the gameplay in the process.

The catch is that any vehicle could push any other vehicle, i.e. a Kublewagon could push a King Tiger off the road just as easily as the other way around, so in practice the result was not much different than what we have now in CMBN.

Exactly.

Steve

[James Crowley]

OK Steve, I stand corrected and hope for extremely small improvements over time.

And thank you for answering my query/complaint in such fulsome fashion.

I shall go back to loosing my game with the edge-hugging 'magic' gapper