Walls, Windows and other assorted MOUT problems

[undead reindeer cavalry]

You can see the trenches from all over the map. The whole defense zone. You can see precisely, without even sending in an observer, all the fallback routes that have been planned by the defender. That's realistic? Maybe it is for the U.S. beating up Syria in 2008, with UAVs and thermal imaging. For the U.S. trying to beat up Germans dug in in Normandy? Not so much.

it's not realistic for 2008 either.

[Wildman]

Perhaps a building that only fits over the trench mesh could be made that is placable by the user...This shows where the unit actually took the time to build/dig/cover themselves. Heck then you could have trenches that are really just ditches and then part of the ditch that was improved defensively.

Not sure how easy it is, but if it acts like a unit then the FOW affects it...the hard part would be coding in the placement restrictions.

[TOG]

I don't know if it has been suggested (probably yes), but is it possible to make a trench unit, which infantry would enter like a transport vehicle? It could be made very hard to spot (harder than infantry in the open) and it would give some protection. Enemy would have to spot it. It could be more complicated to model long trenches, but one squad fighting positions can be created easily just like bunkers in CMSF now, just very flat and hard to spot.

edit: just read some more of this topic, sorry for writing sth that was already mentioned

[undead reindeer cavalry]

having 2 different worlds to keep track of and reconcile is bad enough, but 48 or so? I'd go so far as to say impossible.

what a weird comment.

is that comment accurate or just something you happened to write?

did Charles tell you something like real trenches would require a separate copy of the terrain mesh for rendering the world to the player?

and now you interpolated on that and thought it would mean that in coop there would be a separate copy to be kept for each individual player?

the problem behind this stuff is that CMx2 would require a separate copy for the player?

you need to keep track of spotted units separately for each player anyway, so that is not the problem.

most likely, then, the problem is that you would need separate copies of the terrain mesh for both rendering (the world as shown to player) and physics calculations (used by the game under the hood).

CMx2 would have real time physics engine directly tied to the terrain mesh.

thus you could not have non-spotted trenches because it would cause the game to miscalculate stuff. an enemy soldier in a non-spotted trench would cause the soldier to be within ground -- the physics engine would not allow it.

you could have it only by keeping a separate copy of the terrain mesh for under the hood real time physics calculations and keeping another copy for rendering the world to the player.

so, CMx2 could not have stuff like real trenches because of stuff like real time modelling of vehicle suspension?

wow.

if true, color me impressed.

[Thomm]

You could have it only by keeping a separate copy of the terrain mesh for under the hood real time physics calculations and keeping another copy for rendering the world to the player.

so, CMx2 could not have stuff like real trenches because of stuff like real time modelling of vehicle suspension?

wow.

if true, color me impressed.

From what I understand, this appears to be the core of the problem. On the other hand, the terrain mesh for rendering changes all the time, anyway, because of terrain level-of-detail, so there has to be a different representation of the terrain internally???!?

That last remark of yours, was it meant to be sarcastic? I truly cannot figure it out.

Best regards,

Thomm

[Thomm]

You could have it only by keeping a separate copy of the terrain mesh for under the hood real time physics calculations and keeping another copy for rendering the world to the player.

so, CMx2 could not have stuff like real trenches because of stuff like real time modelling of vehicle suspension?

wow.

if true, color me impressed.

From what I understand, this appears to be the core of the problem. On the other hand, the terrain mesh for rendering changes all the time, anyway, because of terrain fog-of-war, so there has to be a different representation of the terrain internally???!?

For me the problem is that terrain fog-of-war simply is not supported by the current data structures, otherwise we could have spottable walls, e.g., and since trenches are terrain objects, this will be hard to change in the short run. I do not think that the actual mesh data could not be tweaked as needed (terrain LOD proves this).

That last remark of yours, was it meant to be sarcastic? I truly cannot figure it out.

Best regards,

Thomm

[Thomm]

By the way, did we agree on whether we would accept Borg-spotting for trenches, or not? Would trenches be expected to vanish again as soon as soldiers loose LOS?

Best regard,

Thomm

[Other Means]

Each unit would have to have a copy of the trenches or not, dependent upon whether they had spotted it, hence Steve's figure.

Click on a unit that has spotted the trench and you see what they see - trench or no trench. Unit A may see trenches 1, 2 & 3 while unit B sees trenches 4, 5 & 6. Unit A has a chance to see trench 4 but may not due to smoke etc. Except if they've previously spotted it - in which case the trench would still be shown because it's not like they'd forget it because of some smoke. Unless they get the trench spot passed via C2. So the complexity grows.

The spotted status of the trench has no effect on the physics - HE will still kill or not independent on whether it's been spotted or not.

Personally though I'd accept borg spotting for trenches. Being able to see them from the off now really hurts the game. Of course, they need to be able to be player placed as well...

[Redwolf]

Can't say I see a problem with trenches being subject to Borg-spotting (as long as it isn't pre-spotted).

It's not that you'd sneak up your trenches from the enemy flank later.

The Borg-spotting we had in CMx1 was mostly an issue with unrealistically much return fire coming back at an ambushing unit. Since trenches can't be destroyed by fire and area will would be issued manually by the player I'd even go as far as saying Borg spotting for the trenches doesn't make any real difference.

[Battlefront.com]

URC,

most likely, then, the problem is that you would need separate copies of the terrain mesh for both rendering (the world as shown to player) and physics calculations (used by the game under the hood).

Yes, that's what I mean by "world".

CMx2 would have real time physics engine directly tied to the terrain mesh.

It does and that's why this is an even more difficult issue to tackle. It's not just the graphics, but all the junk that tells CMx2 how to interact with the "world". See next example...

thus you could not have non-spotted trenches because it would cause the game to miscalculate stuff. an enemy soldier in a non-spotted trench would cause the soldier to be within ground -- the physics engine would not allow it.

It's worse than that There are "tables" which exist for the system to quickly, and efficiently, check things like LOS, LOF, pathing, and other things. These are extremely big things from a resource standpoint. Lots of RAM and CPU cycles to create and maintain them. You'd have to maintain the master copy and lists of differences on a unit by unit basis. This would have to be done for each player, which in a 24 or 48 player game would be rather massive in total I am sure.

An alternative, and slightly less intensive, system would allow each player only two views of the world; actual and perceived. Thus if one unit spots something like a trench, then instantly all units in the game for that side would know there was a trench there. It's still a real bugger to do, and it does take up a lot more resources, however it's easier than trying to have FOW terrain be at a unit level. Which IMHO is fine because I think this is the way players would want it anyway.

you could have it only by keeping a separate copy of the terrain mesh for under the hood real time physics calculations and keeping another copy for rendering the world to the player.

Actually, they are tied together in the sense that the terrain is what the physics models use for their own calculations. If a tank is moving along and experiences a bump in the mesh, the bump is interpreted by the suspension physics which then instructs the model how to behave so that it looks like it is going over the bump instead of through it.

so, CMx2 could not have stuff like real trenches because of stuff like real time modelling of vehicle suspension?

Yes, along with trajectories, particle effects, smoke behavior, height information for LOS/LOF, etc., etc.

if true, color me impressed.

I presume the colors of white and blue would be good for you?

Steve

[Battlefront.com]

Sergei's comment about visually covering up things is a good place to recover something I've said several times in this thread:

I don't think it's impossible to do either. It's more a question of how much work, ie. how long time it will take from Charles to implement something.

All of these things we're talking about are possible given enough programming time and computing resources. When we looked into this years ago we decided that we didn't have enough of either to build it into the initial version of CMx2. Here we are several years later and we still don't see us having enough of either to do it.

As I said a few pages before, does anybody know of any 3D game that has FOW terrain that modifies the mesh/physics? I can't think of a single game that does but can think of dozens that don't.

Steve

[Battlefront.com]

Oh, and to get back to something from a few pages ago...

I completely agree that from a game standpoint giving the player the ability to place defensive works, like trenches, during Setup is far more important than having terrain FOW. We've always understood these things to be separate issues and, thankfully, from a coding/design standpoint they are indeed separate things. Placing defensive works involves a lot of UI coding but doesn't impact the "world" once the game start, which means it's a lot easier to do than FOW terrain. That's why one is on the schedule and one isn't

Steve

[Other Means]

Giving the player the ability to place trenches is nullified by them not being hidden to his opponent.

I think everyone would accept borg spotting for terrain - it's explainable in real life by them never moving.

But the ability to hide good cover where the enemy doesn't expect it and can't see it is extremely valuable.

[Battlefront.com]

Other Means,

Giving the player the ability to place trenches is nullified by them not being hidden to his opponent.

Not at all. The primary GAME reason to set up ones own trenches is to set up a defensive strategy that is unique and customized to the overall defensive plan. With fixed trenches, like we have in CM:SF, you are either forced to conform (to some degree) to the scenario designer's defensive concept or to purposefully not use trenches that the game provides you with.

Also, it's faulty logic to presume that having trenches initially hidden will actually mean something for a specific scenario. I've played plenty of CMx1 games where I spot the trenches in plenty of time to react to them. True enough, it's an advantage to know about them before the game, but a good offensive plan will work no matter if this is the case or not. Therefore, practically speaking hidden trenches might not make any difference. Plus, I've played plenty of CMx1 games where I spotted trenches within the first seconds of the first turn. Might as well be Setup Phase since I, as the attacker, have plenty of time to use that information to my advantage.

I think everyone would accept borg spotting for terrain - it's explainable in real life by them never moving.

I agree. Unfortunately, it's not going to happen because it doesn't make FOW terrain any more viable as a feature. It's like being on a fixed income of £30,000 a year and being offered a castle in Cornwall for 75% off the £2,000,000 starting price. It's a Hell of a big discount, but it doesn't make it any more practical to own. Well, at least not since the mortgage meltdown

Steve

[RSColonel_131st]

Not at all. The primary GAME reason to set up ones own trenches is to set up a defensive strategy that is unique and customized to the overall defensive plan.

If the opponent can see my unique positioned trenches, then he can figure out my customized defense plan before it hits him hard. To try something "out of the box" it doesn't help when the other player knows about it immediatly.

True enough, it's an advantage to know about them before the game, but a good offensive plan will work no matter if this is the case or not. Therefore, practically speaking hidden trenches might not make any difference.

I find this reasoning funny. Yes, a good offensive plan will work no matter what. But if you are playing an opponent with a bad offensive plan, it will much more likely work if he knows your setup beforehand - so you are basically granting a handicap to less capable opponents, just because a good opponent would win either way?

The whole point of Human vs. Human play is that one of the two will likely have the less quality tactics. Assuming your opponent will have an infallible attack plan anyway and thus it doesn't matter how well hidden your defense is, is akin to saying you'll lose anyway.

[Taki]

Thats not true Conoel.

You will loose anyway against the overrated US-Equipment!

[Redwolf]

This brings us back to the age old topic of balancing some unrealism with things that might be unrealistic by themselves but even things out to be a more realistic package.

In this case, if you have trenches that are placed in setup but not subject to FoW you could "make good" for it by giving the defender more trenches. Then the attacker wouldn't know which ones are the ones in use and cannot make use of the unrealistic lack of FoW.

[Guest Guest]

deleted per user request

[Redwolf]

Having trenches all over the place is not an acceptable solution, because they change the terrain cover essentially from whatever it was to "dense" no matter what.

I'm not talking *that* many.

Twice the amount needed I'd say.

Yes, that will allow people to slow down vehicle movement in certain areas, but that isn't entirely unrealistic either.

[Guest Guest]

deleted per user request

[Redwolf]

Still no good. What happens when you take fire from a location that only has LOS to one set of trenches? What about 155mm airburst with it's massive area of effect? If you have to pick between one platoon position and another to fire on, it won't take long to figure out which one has the enemy and which one is a decoy. Then what, stuck making squad sized positions with platoon sized decoys?

Well, I'm not saying it's a solution that's anywhere close as good as FoW trenches like in CMBB.

I'm just saying that CM:SF style trenches, not placeable, not FoW, will be catastrophic for a Normandy game.

I have repeatedly stated that I think the loss of graphics quality from using "carpet" like trenches like in CMBB is, as ugly as it might be, the better solution than doing that to the game. In addition, I think that if you assume the placeable, FoW trenches are very narrow then the graphical uglyness is bearable.

I don't propose to get rid of the CM:SF trenches. The scenario designers can place them. Looks good, the defender might use them. If you then give them glorified foxholes in the form of "carpet" trenches you still give the defender the option to put his troops whereever he likes.

This would restore gameplay and not be too outrageously hard to implement. No breakup of the terrain ground 3D model at runtime. And as I said in this thread many pages ago, the carpet trenches are about as difficult to model as APCs. They do essentially the same thing, they hold troops which partially stick out, they offer cover and concealment (troops inside are hard to spot) and APCs and trenches themselves are subject to the same rules of FoW. Since we already have FoW APCs the trenches can share some of the code. And the trenches don't move.

It's just getting ridiculous imo. No matter what, if players can see distant trenches, the multiplayer game is pooched. The only way to fix it is to restrain the camera so they cannot be seen or correct the engine. Or, play with opponents who vow now to look beyond the view of their men. (And even then, you still get to see muzzle flashes no matter what...)

I have given up on it personally, and I'm a lot happier just playing CMx1. I like CMSF's visuals better but the gameplay is still years off.

The point is, it doesn't have to be. You are measuring me by the wrong standards, I'm just exploring how we can get useful defenses without getting Charles into a mental institution.

%%

Do I think this issue makes or breaks CM:WW2?

Kind of, yes.

A nasty surprise defensive setup on part of the German player, in Normandy in particular, is at the centerpiece of many of the most enjoyable CMx1 games I had. The creativity to design that defense to be surprising and effective is one of the best ways to spend a mentally relaxing evening.

You can't have that gameplay experience if there is no cover for the defender that isn't pre-spotted before the game even begins.

And then the attacker will counter your setup and the defender's gotta move. No moves without trenches, that's why they invented them in the first place.

%%

To conclude praying to the mostly converted, I'd like to repeat my point about reviewers. You release a Normandy game with only pre-spotted defensive positions, you'll never get rid of the image even if you correct the mistake later.

I know everybody wants the thing released earlier, the company in particular.

But taking that risk to not live with the carpet trenches? Is that worth the risk? CM:Normandy is not going to graphically compete with high-budget WW2 titles anyway in the graphical department, you compete in the realism and gameplay department.

[missinginreality]

The only way to fix it is to restrain the camera so they cannot be seen or correct the engine.

Adam you raise an interesting point here; in Medieval Total War 2 that is how the camera works; restrained to within a certain distance of your units; from high up one can see the whole area but the closer the camera gets to the ground the more restrained it becomes; it is a nice system.

[Battlefront.com]

There's really nothing more to discuss. There are some that think this is the most important issue in the whole entire game, and therefore nothing short of having FoW trenches will make them happy, and then there are the 99% who don't. We've always said we will never be able to make all people happy all the time, so it's absolutely useless to try.

We are not going to have 2D trenches. We are not going to have 3D trenches that are covered by something that makes them invisible to the other side until spotted.

If that's a deal breaker for someone, that's they're choice to make. Somehow I think we'll manage to survive.

Steve

[C'Rogers]

So after taking a few months away from the forum I read this entire thread, think up what I hoped to be a thoughtful post, only to have Steve say what he should have at the beginning. It would be a ton of work for for a feature which, while desirable, is very far away from a crucial component in their minds.

Well actually he did say that, I think. But round and round it went.

[Redwolf]

So will we have user-placeable foxholes?