Misidentified Units Back In?

[Other Means]

Actually this isn't the case, at least as far as I'm concerned.

I think it's more likely that those who used to enjoy it and have moved on to games without it have forgotten just how good that feature is, and the added "immersion realism" that it gives.

Sure, it has some imperfections like the direction facing thing. But having to wonder "gee, I wonder what that thing really is, I need to find out" is a significant and very good piece of the CMx1 gameplay.

I think BFC would be doing themselves a disservice to play it down.

My experience comes from having played CMx1 right up till now very regularly, then trying another WWII game that doesn't have mid-id. It's like OMG, this sucks!! A whole aspect of recon and tension is removed from the game.

While "how should it work with relative spotting?" is a legitimate question, there are perfectly tractable answers.

The more fundamental issue of "it wasn't allowed in the designin beginning and now has to be patched in" is the real killer. Darn shame to hear that really, but c'est la vie.

My main point here is "don't underestimate the power of this feature: it really really affects gameplay in a positive way".

GaJ

Couldn't agree with this more. It's the uncertainty that effects everything that you do. You end up never quite trusting what you see initially and that adds an awful lot to the game.

[Steiner14]

I think it was one of the very fascinating main challenges in CMx1 (playing human opponents, ofcourse), to find a unit deployment that kept high flexibility lasting long into the battle, until all enemy major units were trustfully ID'ed.

[c3k]

FWIW, I understand the desire to have misidentified units in the game. I have no idea how that would play. As stated above, if a PzIV is misidentified as a Tiger, that information disseminates through command links. When the initial unit correctly ID's it as a PzIV, how would that play? One unit "sees" a PzIV, another "sees" a Tiger. Remember, what you see is how it works. Targeting, penetration, hull down status, etc.

In my dream world, I'd like <?> contacts to go through several levels of identification. I'd think an initial <?> would swiftly be discriminated into one of three types of <?>; infantry (men), wheeled vehicles, or tracked vehicles. Icons of men, trucks, or tanks with a "?" over them would convey that information. Next, as the <?> gets resolved, you'd see generic men, vehicles, or tanks as icons. At this point you would have NO idea what unit it was or where in the command hierarchy it was. (Right now in CMSF as soon as you resolve a <?> into a solid contact you get its exact name and type, i.e., "3rd Platoon HQ".)

I have no idea what it would take to code in layered information.

Note that this is a fundamentally different approach from MIS-identifying units. Instead it is a way of UN-identifying units. There would be no false information disseminated, rather just a lack of information.

Further thoughts: "Men" would cover any non-motorized unit, including mortars, guns, MG's, etc. The movement of the crew would be what is seen, not necessarily the weapon. Later, as identification improves, then the type of unit would be seen.

Any non-moving vehicle would be identified by its engine noise. Hence, either icon, truck or tank, could be used. However, once a vehicle is moving or in sight, any tracked vehicle should use the generic tank icon. (The noise of tracks is so distinct as to preclude misidentifying a tracked vehicle as a wheeled vehicle. I would classify initial contacts with half-tracks as a tracked vehicle.)

An example of the layers, using a PzIVH would be:

<?>: Possible Location of an Enemy Unit

<?+Tank Icon>: Possible location of Enemy Tracked Unit

<Tank Icon>: Enemy Tank Location

<PzIV> : PzIV Location

<PzIVH>: Specific model identified as PzIVH

<PzIVH, Platoon HQ>: Self explanatory, but specific platoon unknown (tank number gives that information, so perhaps include it)

Next would be status information, ie, if the the barrel is bent, then the allied player would know that the main gun is inop, or maybe they would know if it's immobilized.

This may be too much.

Thoughts?

Ken

[Lanzfeld]

Sounds excellent to me. ^^^THIS^^^

De-identify...not misidentify....I like it.

[Flanker15]

An example of the layers, using a PzIVH would be:

<?>: Possible Location of an Enemy Unit

<?+Tank Icon>: Possible location of Enemy Tracked Unit

<Tank Icon>: Enemy Tank Location

<PzIV> : PzIV Location

<PzIVH>: Specific model identified as PzIVH

<PzIVH, Platoon HQ>: Self explanatory, but specific platoon unknown (tank number gives that information, so perhaps include it)

Next would be status information, ie, if the the barrel is bent, then the allied player would know that the main gun is inop, or maybe they would know if it's immobilized.

This may be too much.

Thoughts?

Ken

CMx1 games had this or something very close to it I'm pretty sure. As well as misident.

[SlowMotion]

In my dream world, I'd like <?> contacts to go through several levels of identification. I'd think an initial <?> would swiftly be discriminated into one of three types of <?>; infantry (men), wheeled vehicles, or tracked vehicles.

...

Thoughts?

I'd also like to see more levels of identification. IMO the most detailed info especially about infantry units should be available only from short range. And inexperienced units might not get detailed ID from things like vehicles even then. So instead of PzIVJ they might think it's a PzIV or just a tank. Or instead of Sdkfz 251/1, a rookie unit might know it's a halftrack but not have more accurate info.

[panzermartin]

Yes, I doubt the average GI could tell the difference between a PzIVH and a J even from point blank. Maybe (in elite/iron levels) some things should not be revealed till the battle is over and you can overview the battlefield.

[Battlefront.com]

I don't think it's entirely fair of you to say that.

Considering how often we are told that old features are not on the menu now but are on the mythical 'to do' list (that must now be large enough to cause a slight wobble in the earth's rotation) I don't it is entirely irrational.

The list is, indeed, large enough to create anomalies in the space time continuum. For example, I typed this message in 2026.

It's not wildly optimistic to expect features that were in the series 10 years back to be included. Especially not given the spectacular delays that inevitable do raise expectations.

Which is the point I made earlier. It is wildly optimistic to expect all the old features AND all the new features AND to have it all delivered in 1/2 the time it took us to make any previous CM game. Something has to give, and sometimes that is a feature from CMx1 we wanted to bring forward, sometimes it is a new feature that none of you have even though of but would kill for if you knew about it. If you guys knew all the new stuff that has to be cut for a game to be shipped your heads would probably explode

Also, Kubelwagens mistaken for Tigers, it's hard not to laugh at that charaterization of the ID system of CMx1. It was nowhere that bad.

True, it wasn't. But sometimes it's useful to use extreme examples because it makes the essence of the issue more apparent. And in this case it is actually not something that should be laughed at. Getting a system in place that DOESN'T have a Tiger being mistaken for a Kübelwagen takes programming time.

The point I was making, however, seems to have been lost because of the example I gave. Which is:

"Which means we have to program an "alternate reality" within the game so that when you think you see a turreted tank and it turns out to be a StuG that the game is clear on the rules for both players. It's not impossible to do, of course, but it is quite a bit of work none-the-less."

Take the Kübelwagen/Tiger 1E out of the example if you like, but the technical hurdles for us remain the same.

Steve

[Battlefront.com]

The system we want to implement is not dissimilar to what c3k outlined. It helps with the "alternate reality" problem a great deal because the 3D representation is shown only it is positively identified as a particular type of vehicle. There are still issues with things like skirt armor, but those problems are far more isolated and simplified than trying to have a PzIV mis identified as a StuG. In other words, it makes the programming of the alternate reality more simplified.

The resulting system won't be as visually interesting as CMx1, but it will probably be more realistic and at the same time easier to implement.

Still, nothing we can just pop into the game and therefore it won't be in CM:BN.

Steve

[stoex]

The system we want to implement is not dissimilar to what c3k outlined. It helps with the "alternate reality" problem a great deal because the 3D representation is shown only it is positively identified as a particular type of vehicle. There are still issues with things like skirt armor, but those problems are far more isolated and simplified than trying to have a PzIV mis identified as a StuG. In other words, it makes the programming of the alternate reality more simplified.

An interesting bone, Steve...while I have no idea, obviously, of how this is eventually going to work when it gets into the (or better a) game, your statement raises some interesting questions about how such an implementation will affect the AI in particular.

Since I know there is no point in my asking these questions now, I will save them for an appropriate time. Just chipping in to say my imagination is running wild.

[ChrisND]

I'd also like to see more levels of identification. IMO the most detailed info especially about infantry units should be available only from short range. And inexperienced units might not get detailed ID from things like vehicles even then. So instead of PzIVJ they might think it's a PzIV or just a tank. Or instead of Sdkfz 251/1, a rookie unit might know it's a halftrack but not have more accurate info.

This is something that has irked me for a long time. All infantry should be generic infantry contacts until they do something to betray their role. I bug Steve/Charles about this about once every three months. Speaking of which, that three months is up, time to make a new post...

[GreenAsJade]

The posting about "at least display them as ? till we know what they are" triggered a sudden realisation for me why this is hard in CMx2, compared to CMx1.

In CMx1, the unit was just a point on the map, as far as the display was concerned. A type, a coordinate and an orientation, probably: very abstract. The rendering of it could take that information and display whatever it liked: decoupling the visible geometry of the unit from its behaviour.

However, in CMx2 it suddenly clicked for me that the behaviour of a unit is tied to its geometry. For example, we read that the trajectory of projectiles is plotted, and displayed (?), and its actually calculated which surface of the target it interacts with.

So the rendering of things is intimately tied with it's interactions/physics.

Just a guess. But I can see how that makes it hard to change the rendering based on identification status....

GaJ

[Elmar Bijlsma]

Ah, but ordinance falling on something you haven't spotted will still hit it. So to the engine there already is a difference between what is shown and what is there to be interacted with. One would think that ability can be expanded upon.

[Battlefront.com]

GaJ and Elmar are both right.

Currently the game system simulates both sides as if there is no FoW. This means as far as the game is concerned all the geometry is there and moving about even when one person can't see it. The actual shaded polygons displayed to the user are irrelevant to the system because it can resolve the geometry without them. Kinda like playing the game with the monitor off. The game doesn't care because it is simulating all those polygons and intersections whether you see them or not.

So in that sense Elmar is correct. But it's also where Elmar is incorrect

The problem, and where GaJ is correct, is that the shaded polygons that you guys do see are literally where the geometry of that object is. In other words, if you see the top part of a vehicle behind a house, that's exactly where it is. It can't be somewhere else because it isn't somewhere else. It's right behind that house. While the system can turn the polygons on/off as needed by FoW, it can only ever turn them on where the object actually is at the time the polygons are made visible.

The game has a definite sense that Soldier in exactly this spot doing exactly a very specific action. It is not in another spot, it is not doing a different action. If we portray things to one player in a place or situation that is NOT "real", then we have the problem of needing to track and resolve *two* realities. One that is "real" and one that is on some spectrum of real/false. This is something the game is not equipped to handle and is difficult to handle because of all the additional overhead, visual issues, and potential for a lot of bugs/gotchas that crop up as a result.

This is definitely an area where a highly abstracted system, like CMx1, has it much easier than CMx2. Since no system can possibly be perfect, it's a tradeoff between pros and cons. The pros of CMx2's more literal system have very little cons when compared to the alternative CMx1 abstraction system. Plus, we can work around this problem given enough time and energy, it's just a bit more difficult.

Steve

[Taki]

So we have the same Everything or Nothing Spotting System like CMSF? That was one of the worst thing in this Title.

It was so well praised and it never lived to its expectations. The US Forces was still the Borgspotter. When a M1A1 spotted an Enemy Tank a Milsec later the whole army knew where it was.

I liked the System overall. But the Times for other Units knowing where the Enemy is was Waaaaay to short.

So we have the CMSF Spotting System in?

[Wiggum]

With the CMSF Spotting System as soon as a unit fired there was a really big chance that this unit get spotted directly.

At not as a imprecise contact...

I can only remember a few situations where i was not knowing where the fire was coming from.

At least you could always track the fire watching at the tracers...which was a bit like cheating.

I think Such "WTF where is this fire coming from" situations should really increase in the Normandy Game.

And i hope that those tracer fire stops (i would say, keep it realistic), this made it far to easy to spot the location of any enemy unit.

[stikkypixie]

So we have the same Everything or Nothing Spotting System like CMSF? That was one of the worst thing in this Title.

It was so well praised and it never lived to its expectations. The US Forces was still the Borgspotter. When a M1A1 spotted an Enemy Tank a Milsec later the whole army knew where it was.

I liked the System overall. But the Times for other Units knowing where the Enemy is was Waaaaay to short.

So we have the CMSF Spotting System in?

I am quite sure the US Army in 1944 will not be able to borg spot. Just call it a wild hunch .

[Michael Emrys]

The game has a definite sense that Soldier in exactly this spot doing exactly a very specific action. It is not in another spot, it is not doing a different action. If we portray things to one player in a place or situation that is NOT "real", then we have the problem of needing to track and resolve *two* realities. One that is "real" and one that is on some spectrum of real/false. This is something the game is not equipped to handle and is difficult to handle because of all the additional overhead, visual issues, and potential for a lot of bugs/gotchas that crop up as a result.

Ahhhh, the meaning of your earlier comment about "alternate reality" is now more clear.

Michael

[born2lose]

So we have the same Everything or Nothing Spotting System like CMSF? That was one of the worst thing in this Title.

It was so well praised and it never lived to its expectations. The US Forces was still the Borgspotter. When a M1A1 spotted an Enemy Tank a Milsec later the whole army knew where it was.

I liked the System overall. But the Times for other Units knowing where the Enemy is was Waaaaay to short.

So we have the CMSF Spotting System in?

Not having been in the Army I am not sure whether i am right here, but surely now a days with modern comms if a US tank did spot an enemy tank it probably wouldnt be long before the whole army knew where it was, although it might take a bit longer than a mil sec to do it in real life i am sure within the scope of CMSF all your forces in the game would def know about it very quickly.

[Taki]

Im not question that it is fast. But a bit too fast simulated in the Game. Same for the Nato Tanks. I know they have their Spotting Systems and Electronic Maps where they can place that Contacts. But in the Chaos of Battle its a tad overmodelled.

Its that nice thing "on Paper" looking that good not working in reality that well. Take a look at "Ambush Alley" and you see what happens to that Fancy High Tec Armys in Action. ;-)

[GreenAsJade]

Who cares whether instant id and borg awareness is realistic in the modern milieu or not? That's a CMSF topic.

The point is that in WWII it is both realistic (that that enemy units take time to identify) and a desirable game play feature. It strongly affects tactics. Remember the mission of BFC: to keep our strategy wargame experience alive? Even a WWII board game can have mechanics for unknown precise identity (in an honesty system) and it is a desireable feature. It's a bitter irony that the more realistic implementation of the game has made this harder for BFC to add in, c'est la vie.

GaJ

[Battlefront.com]

True, no system is perfect and each has it's own difficulties. Theoretically it was much easier to add pretty much anything to the CMx1 system since it was so abstract. But over time the game engine became so difficult to manipulate because abstractions on top of abstractions starts to produce it's own set of problems. Those problems then have to be worked around and that usually creates more problems. Which is why the CMx1 game engine was dead even before CMAK was released (note how few new features were in that game).

On top of abstract systems becoming more problematic as they become more complex, the results will always be limited by the foundation under the new feature. It's like taking a 1 story house built on a sand dune and trying to make it into a 10 story hotel. It just won't work, therefore if that's what you want to build on that spot you're either out of luck or you need to tear the whole thing down and start from scratch. The realism fidelity of CMx1, therefore, was inherently unable to be improved.

The downside of a much more literal game engine, such as CMx2, is simple things often take longer to add. Sometimes a LOT longer. But the progress is sustainable and once a feature is added it is unlikely to prohibit some other feature from being added later on. Which in turn means that the day when we have to chuck CMx2 aside for CMx3 is not even on our radar after 3 major releases, where as we decided to chuck the CMx1 code while the 2nd one (CMBB) was still in development.

For us, and for you guys, there's no question at all that things are far better now than if we had simply thrown together another short lived, abstract engine.

Steve

[GreenAsJade]

Woo hoo

GaJ