2nd Expansion Same Old Questions

[Lethaface]

I would rather have a shopping spree in a place with lots of everything!

[Kat Johnston]

There is no chance of us doing a random map generator. Zero. What we are instead doing is combining CMx2's existing ability to use premade maps with a new system which allows smaller premade "Mega Tiles" to be assembled on the fly into new maps. This is a system which borrows from board and miniature wargames. User made Mega Tiles will increase variety even more.

I'm very pleased to read this. If the tiles are small enough, I'd say you could call it a random map generator (UFO/XCOM springs to mind as an example of this being done well; it really helps for replayability).

[Paper Tiger]

This Mega Tile concept is very, very exciting. Designing detailed maps is often an enormous amount of work/fun. I love the idea of creating 20-30 very highly detailed tiles for QBs in the future.

[TempV]

The solution we have combines the ability to Cherry Pick with the ability to still have structured forces. It should address most of the concerns anti-Cherry Pickers have while at the same time allowing for a similar degree of historically impossible forces as was possible in CMx1. At the same time it will work with the underlying C2 model and historical plausible forces.

That's very good intention. I recall Flames of War (tabletop wargame) system where a player must have some "core" troops in points-based limit and he has some freedom choosing support troops. For example, a player fielding Infantry company has to place Company HQ and 2 rifle platoons. Then he may add another rifle platoon or have some weapon platoons (heavy MGs, mortars) and division support troops (tank platoon, assault gun platoon, field gun battery, air support etc.) provided he fits the point limit. It usually equals 1500 or 1750 points which mean core troops plus 2-3 weapon and support platoons.

What we are instead doing is combining CMx2's existing ability to use premade maps with a new system which allows smaller premade "Mega Tiles" to be assembled on the fly into new maps. This is a system which borrows from board and miniature wargames. User made Mega Tiles will increase variety even more.

This is good feature too, I'd like to see it in computer game.

Personally, I don't consider CMSF QB system as bad as it's often spoken about althought it has some obvious limits. With some tuning (and luck ) one can get pretty balanced battle. For example, putting SBCT vs Respublican Guard or SF with minus 20-30% correction to Blue side or +25-40% to Red one in the case they are attackers, will produce a very challenging battle for Americans. I recall you've said about minor fixes for current QB system planned in 1.20. Will they see the light and deal with some strange selection issues (for example, only AT assets), esply for tiny and small battles?

[Battlefront.com]

The Mega Tile concept is certainly one that is proven viable in many games, though mostly non-computer based. There are a few tricky things which we've got to work through to ensure that user created Tiles can be correctly used by the game. With a board or tabletop game, the environment is carefully controlled either by game developer created map boards or direct player input on map feature locations.

TempV brings up the main dilemma CM has with force choices. Most wargames out there have fairly limited unit variables, like Experience, Condition, Equipment, and even TO&E in general.

For example, Close Combat had very little sense of TO&E. If you asked for an infantry based force it would just stuff a bunch of infantry teams and some heavy weapons, plus a support unit or two, and bingo... you got a viable force. This is similar to how CMx1 worked.

But CMx2 cares, a lot, about organization. The picking system can't "Cherry Pick" below the Platoon level. If the battle is small then there are very few actual selections to be made before it feels the weights are roughly equal. But what the heck is "equal"? That depends HEAVILY on things like what the other guy has, the terrain, quality of the soldiers, quality of the equipment, weather, amount of time in the battle, what type of battle it is, etc., etc.

It's the same problem we face with a true random map generator. The game elements which make the game deep and rich for a Human scenario designer are the same things which make any sort of automated system extremely difficult. Which makes perfect sense. If I tell you that you have only a choice of vanilla or chocolate icecream, and that you can have it either in a bowl or a cone with between 1 to 3 scoops, well... it's a pretty easy thing for us to design a system which can probably figure out what you want and make that choice. But if you instead have a choice between 50 flavors, 10 different size/type of containers, 30 different topping choices, different ways of measuring the serving size, etc.... well, it's going to be pretty tough to get you something you'll be happy with each and every time you ask "I'd like a medium amount of some type of icecream with chocolate in it with a few toppings and in a cone of some sort. Oh, and I want it to be exactly what I'm picturing in my head". Yup, really sucks having all those variables

As I've said from the beginning, the primary problem with the CMx1 system in CMx2 is that Cherry Picking isn't inherently workable because CMx2's entire game structure is based around realistic military organizations. Free-form Cherry Picking simply doesn't work with that type of system. Or at least it only allows for an unrealistically organized force mix, which we still think is a crappy solution if that's the only solution. As it was CMx1 barely pulled it off and it had a lot less variables and concerns to work around.

Yup, this is definitely not an easy thing to fix. Which is why we have a large block of development time scheduled for this one feature. There are no quick fixes here.

Steve

[[hirr]Leto]

I think the cherry picking element was more useful, fair and fun at the human vs human level (in CMx1). Do you see the major problem here as trying to set up a QB system that is to be used with human vs AI games in CM2?

I get that the now unit organization element is going to be predominant, but unless there are at least a good number of different unit organizations to choose from (battalion, regiment, company and platoon), the QB system may get a bit ho hum... especially for guys like me who think variety is the spice of life and the most fun in picking forces in QB's is your ingenuity in putting together unique and effective forces, from the battalion pick all the way down to the single unit pick.

I am very much looking forward to seeing how you guys will be doing this, but admit, I am a bit tremulous about the level of flexibility QB players will have in choosing their kits. But then again, I guess we all have to retrain ourselves to the fact that there will not be a very large number of individual units (such as tanks, vehicles, arty, support weapons and troops) to choose from in the context of CM Normandy...

Cheers!

Leto

[Scipio]

I always saw the CMx1 cherry picked forces as some kind of 'Kampfgruppe' meaning 'let's gather what's around here' concept, with somebody who took command, or was ordered to do so. Hard for me to say how realistic this is or was, since my memories are a crude mix of warbooks and movies, both documentary and fictional.

I won't put a spoke in your wheel, Steve. Maybe I missunderstand the basic problem of C&C and force picking.

[Battlefront.com]

Leto,

I think the cherry picking element was more useful, fair and fun at the human vs human level (in CMx1). Do you see the major problem here as trying to set up a QB system that is to be used with human vs AI games in CM2?

No major problem at all. In fact, the idea is for there to be a free-form Cherry Picking option. We have to write code to figure out how to organize them (otherwise Relative Spotting won't work!!!) in some sort of sensible manner, but we can do it even if it is rather silly.

I am very much looking forward to seeing how you guys will be doing this, but admit, I am a bit tremulous about the level of flexibility QB players will have in choosing their kits. But then again, I guess we all have to retrain ourselves to the fact that there will not be a very large number of individual units (such as tanks, vehicles, arty, support weapons and troops) to choose from in the context of CM Normandy...

Oh, there will be Sure, not as much a was in CMBO, but still a ton of different stuff. Just look at how many things are available in CM:SF in its basic form even. On to of that add the Marines, British, and (eventually) NATO stuff. The number of types of forces and equipment to choose from his rather huge. I know because I've had to hand code all of that TO&E

Scipio,

I always saw the CMx1 cherry picked forces as some kind of 'Kampfgruppe' meaning 'let's gather what's around here' concept, with somebody who took command, or was ordered to do so. Hard for me to say how realistic this is or was, since my memories are a crude mix of warbooks and movies, both documentary and fictional.

That's the most common justification for it, true enough, but it doesn't stand up to scrutiny because the player *isn't* just getting whatever happens to be around. Instead he's Cherry Picking until his points run out. "Ooo... I have 15 points left. Let's see, I can get a couple of LMG Teams or perhaps one small AT team. Well, we agreed to play on an open map so I think the LMGs will be a better pick". That's not how Kampfgruppen were formed

Maybe I missunderstand the basic problem of C&C and force picking.

The primary problem is that spotting information, the heart and soul of Relative Spotting, is passed along units according to chain of command. No force in the world would purposefully go into battle without a chain of command, therefore going into a game with perhaps an intact platoon or two and the rest of your force being unassociated would be a major problem. Of course, we could have allowed this and just said "that's what you get for Cherry Picking", but we figured that would be an unacceptable answer.

Steve

[Thomm]

Somehow, I suspect that the whole cherry picking problem is mostly an interface issue. Theoretically, it should not be a problem to present the player with an OOB consisting of empty slots forming a tree-like structure, where he could drag and drop appropriate troop types into.

Also, regarding Relative Spotting, do I recall correctly that it is mostly for presenting the player with Question Markers, with no direct impact on spotting or TacAI?

Best regards,

Thomm

[Wengart]

I've seen my BMP's orient their guns onto question marks as they approach. So I am pretty sure it does effect the AI's spotting capabilities, at least the fairly recent ? do.

[Battlefront.com]

Thomm,

Somehow, I suspect that the whole cherry picking problem is mostly an interface issue. Theoretically, it should not be a problem to present the player with an OOB consisting of empty slots forming a tree-like structure, where he could drag and drop appropriate troop types into.

Absolutely. This is a straight forward problem to solve. The thing was we didn't think we needed to spend the time on this feature at the expense of other features. There's only so many hours in the day, so if something takes x time to do and the schedule is full, x amount of time's worth of stuff must be removed from the schedule. CM: Normandy could have other significant new features in place if we didn't have to redo the QB system.

Also, regarding Relative Spotting, do I recall correctly that it is mostly for presenting the player with Question Markers, with no direct impact on spotting or TacAI?

Not true. If something is spotted, and that information is passed around to a unit, the TacAI for that unit uses that info. What the TacAI for a unit can't do is act on information that isn't directly available to it, which is something a Human player can do. For example, 1st Squad, 2nd Platoon, C Company is the only one that spots some infantry in the open. If the player is an AI player then only 1st Squad can directly do something with that information. If it is a Human Player he can do things like fire blind Indirect Fire into the area, tailor maneuvers to counter that specific threat (e.g. moving an AT asset vs. an anti-infantry asset), etc.

Steve

[Thomm]

Steve,

My bad, memory failed me:

Link to most detailed post regarding question markers, ever.

However, your statement:

If something is spotted, and that information is passed around to a unit, the TacAI for that unit uses that info.

... sounds stronger than what you sketched out in the post above (TacAT only knows that "something is there")

Anyway, I apologize for sewing confusion.

Best regards,

Thomm

[Scipio]

Scipio,

That's the most common justification for it, true enough, but it doesn't stand up to scrutiny because the player *isn't* just getting whatever happens to be around. Instead he's Cherry Picking until his points run out. "Ooo... I have 15 points left. Let's see, I can get a couple of LMG Teams or perhaps one small AT team. Well, we agreed to play on an open map so I think the LMGs will be a better pick". That's not how Kampfgruppen were formed

Steve

Well, you are right, in the war movies they usually didn't go shopping . Anyway, as I have written above, I have no problem with auto purchase if the problems with it will be resolved. I'm especially worried about the module incompatibilty as I described here : http://www.battlefront.com/community/showthread.php?t=87759

[Battlefront.com]

Thomm,

... sounds stronger than what you sketched out in the post above (TacAT only knows that "something is there")

Nope, it's the same thing If the information passed to a unit is a "?" then it only knows "something is there" even if other units know exactly what it is. If the information is instead passed exact information then it knows "exactly what is there". Either way the method of information transfer is accomplished the same way, it's just that the information quality is variable.

Anyway, I apologize for sewing confusion.

Drink a good beer for me and we'll call it even

Scipio,

Ah... sounds like a check isn't being preformed when it should be. I'll make sure Charles knows. You're not supposed to be able to do what you did.

Steve

[Thomm]

If the information passed to a unit is a "?" then it only knows "something is there" even if other units know exactly what it is. If the information is instead passed exact information then it knows "exactly what is there". Either way the method of information transfer is accomplished the same way, it's just that the information quality is variable.

Steve,

That sounds great! Do you have any plans on adapting the UI in a way that gives the player information about the "quality" or "content" of the Question Marker?

Best regards,

Thomm