Steve's recent CMx2 Bones

[aka_tom_w]

this is still a VERY significant bone:

"Battlefront.com

Administrator

Member # 42

posted May 29, 2005 05:16 PM

We do have to be quite careful in the subject matter and scope we choose for each title.

We want to make sure that you guys get something that is still fun to play by the time the next module or full title comes out.

If we start out with an idea for the subject and scope that later on in development seems to be too narrow... then we'll widen it out a bit.

But the days are gone when we'll release a game that people can play for 2-3 years without getting bored of it. Fortunately, with quicker release schedules the need for a game that has 36 months of playtime in it is made unnecessary.

As for the modules working with each other... yup, that is the thoght. Let's say you get a WWII ETO game that is limited to Battle of the Bulge, US vs. Germans (specific unit types). 6 months later we release a Normandy modlule that allows you to play US vs. Germans (specific unit types). The latter will simply add options to the game you already own, kinda like CMBO. Another module comes out that supports Commonwealth for both Normandy and Battle of the Bulge. Now it really is like CMBO. Yes, it took three releases to get it to that stage, but in the mean time other games have been released and others started."

[MikeyD]

A hIdden double-reverse bone?

Considering BFC's steadfast refusal to divulge where the next game's set, wouldn't their line "...let's say you get a WWII ETO game that's limited to the Bulge..." therefore preclude the chance the first game'll be an ETO Bulge game? Like if they had instead said "...let's say get a Vietnam game that's..." we'd assume that using this as an example means we WOULDN'T be getting a Vietnam game.

Next topic: How many angels could dance on the head of a pin, knowing what we know about angels.

[aka_tom_w]

I Wonder how many of these T72 Balkan's on Fire (Latest PC ONLY Gone Gold, game not shipping yet) features we will see in CMx2???

• More than 70 types of buildings and installations that can be damaged or destroyed

•Over 40 types of plants, trees and grass sway in the breeze

High-quality modeling of water surfaces and amphibious vehicle capabilities

•5000+ polygons per vehicle including detailed bump mapping

Scalable realism settings

•Powerful mission, map and structure editor available

Local Area Network (LAN) multiplayer capable

Advanced Physics and Environmental Game Engine

•Vast open spaces with draw distances of more than 2 km

•Topographic height differences on maps of up to 800 m

•Realistic passage of time transitions (day changes to night)

•Dynamic weather and environmental effects: smoke, fog, rain, rainbows, dust, wind effects (I am looking forward to Rainbows in CMx2! )

•Deformable terrain - shell holes and trenches dug out in real time

•Destroyable structures, vehicles and vegetation

•Clouds cast shadows

•Optical lens flares and sun glare effects

•Realistic ballistics – wind effects, shell aerodynamics and flight characteristics, barrel wear and droop

•Shell impact angle calculations

•Detailed armor penetration results

•Post-penetration shell path and damage assessments

•High Explosive fragmentation dispersal calculations

•Realistic vehicle and structure collision detection and damage

•Individual internal and external vehicle system damage modeling including track damage, engine overheats, fuel leaks, turret jams, crew casualties, bailouts and more

QUESTION:

Did they are did they not say something about the possibility of a MAC version of this game?

wondering

:confused:

-tom w

[Gpig]

I think I read early on that there would be NO Mac version of T-72.

happy to be corrected, however.

Gpig

[Panzer76]

Originally posted by Gpig:

I think I read early on that there would be NO Mac version of T-72.

happy to be corrected, however.

Gpig

Join the dark side, muhahaha!

[V]

Originally posted by flamingknives:

No, Dr. StrangeV, I'm sure that you were only obeying orders.

[Gpig]

Panzer76 wrote:

"Join the dark side, muhahaha!"

NNnnnoOOOOOOOOOOOooooooooo!

[hoolaman]

Originally posted by aka_tom_w:

I Wonder how many of these T72 Balkan's on Fire (Latest PC ONLY Gone Gold, game not shipping yet) features we will see in CMx2???

[snip]

I really hope CMX2 gives a much better overall "impression" than T72. Some details like trees and lighting do not look "just-right" to me.

The most realistic and impressive feature in T72 is the destructable buildings, which I hope is in CMX2. Most of the rest of the terrain and environmental features have already been stated as being "in" for CMX2, but I really hope and expect BFC to do it better.

[Michael Emrys]

Originally posted by aka_tom_w:

•Optical lens flares and sun glare effects

This adds nothing to the game for me, and indeed, from the point of view of anyone not looking through an optical device isn't at all realistic. I make some exceptions for sun glare when a rising or setting sun behind enemy units would make them harder to see under many conditions.

•Realistic ballistics – wind effects, shell aerodynamics and flight characteristics, barrel wear and droop

This is a game factor only if the player is acting as a gunner. Not a game factor in a game such as CM. Should be one of those things calculated "under the hood".

Michael

[aka_tom_w]

OK to both of those points....

Good Thinking

-tom w

Originally posted by Michael Emrys:

</font><blockquote>quote:</font><hr />Originally posted by aka_tom_w:

•Optical lens flares and sun glare effects

This adds nothing to the game for me, and indeed, from the point of view of anyone not looking through an optical device isn't at all realistic. I make some exceptions for sun glare when a rising or setting sun behind enemy units would make them harder to see under many conditions.

•Realistic ballistics – wind effects, shell aerodynamics and flight characteristics, barrel wear and droop

This is a game factor only if the player is acting as a gunner. Not a game factor in a game such as CM. Should be one of those things calculated "under the hood".

Michael </font>

[aka_tom_w]

OLD bones from Jan 2005 revisited and listed here

Battlefront.com

Administrator

Member # 42

posted January 31, 2005 04:05 PM

I have to close up this thread due to its size. Feel free to start up another one. However, keep in mind that we (Battlefront) need to design something that is pleasing to both the eye and the grog. We'll be torn to pieces if the 1:1 representation looks silly or woefully incomplete, even if eveything is neato mosquito under the hood. Likewise, grogs will tear us a new one if the important 1:1 simulation aspects aren't done well enough.

In other words... there is no one right answer to this debate you guys are having. We need to have a balance between the two, and that balance is in part determined by how easy/hard it is to program and/or how well it functions from a performance standpoint (eye candy and underlying sim stuff BOTH!). Not even we have the answers to all these questions at this point, so I might suggest that both sides of this debate keep that in mind. NOTHING is decided except for the fact that we are seeking a balance, which inherently means some grog stuff and some eye candy wish list items won't be happening.

Steve

Battlefront.com

Administrator

Member # 42

posted January 30, 2005 08:48 PM

The 3D environment of CM does offer us some challenges in terms of where to draw the line between a simulation of the individual soldier and larger groups of soldiers (i.e. units). It's something we've all be struggling with since the early days of CMBO's design.

Back in the old days of wargaming you'd have a cardboard chit with some numbers and a shape on it. Nobody called for more than that simply because the system was so abstract probably nobody thought of it (at least not thinking it possibly practical).

The first couple of generations of computerized wargames fell into the same category. As time went on and technology improved wargamers wanted better looking maps, more attractive looking units, and of course more under the hood. But again, for anything but the smallest scale wargames nobody thought about simulating the individual soldier to any significant degree. Until, that is, Close Combat came out.

CC was the first commercial wargame to model the individual soldier in detail and in substantial numbers. And for all its flaws, the game worked very well and people saw the value in having the 1:1 soldier simulation. Then CMBO came around...

Our problem, from the beginning, is our chosen scale. MUCH larger than CC's, yet not so much that individuals ceased to matter. But due to technical limiations we never once thought about doing 1:1 because it simply wasn't possible. However, the desire has always been there, at least to some extent. Now comes CMx2...

What we are doing now is giving the individual soldier a place on a larger scale battlefield (larger than FPS, CC, etc). That is the right thing to do. However, there are limits. We must make sure to not lose sight of the fact that this is a larger scale wargame and not a FPS of even Close Combat scale game. Therefore, when push comes to shove, decisions are made which favor the larger scale wargaming environment. Clutter, unnecessary development distractions, big hits to the CPU for little simulation gain, etc. are all bad things for CMx2.

In short... we know what CMx2 is supposed to be and what it isn't. We have this vision very clearly laid out and will not waiver from it. There will be no mission creep.

Steve

[aka_tom_w]

another old bone from the old 1:1

Battlefront.com

Administrator

Member # 42

posted January 21, 2005 12:15 AM

1:1 representation sure does open up a big can of worms. It's one of the reasons why we did not attempt it for CMx1 (though hardware wouldn't have allowed it anyway). However, as has already been pointed out one should not confuse 1:1 graphical represenation with 1:1 modeling or 1:1 control. Three different concepts.

From a GAME standpoint, 1:1 graphical representation is the most important. If I were to make a Top Ten list of complaints from general gamers about CMx1, this would probably be the #2 complaint (#1 is the lack of ridiculously detailed and textured models) constant throughout all three games.

From a REALISM standpoint the most important thing is the 1:1 modeling. Though it is very difficult to do this without the 1:1 graphical represenation, it is certainly possible to do. We could have had individuals run away from generic 3 man squads or more detailed soldier stats. But without 1:1 representation this all seemed kinda hollow so we kept the level of modeling in line with the degree of visual representation. Now that we are increasing the latter, we will also be increasing the modeling to stay in balance. There will still be abstractions, just not nearly as many as there are now.

The interesting thing is that most "gamers" and "grognards" is that they probably pretty much agree that 1:1 control is undesirable. There is already enough to pay attention to without having to get Pvt. Pyle to move 0.5 meters to the left of the tree he is behind so he can get a shot off. It also turns the focus to individual soldiers instead of units and the formations they belong to.

Obviously, if you are smart you'll see that we're not going to implement 1:1 control, but are going to do 1:1 visual and 1:1 modeling. The results should make the game more fun to play and also far more realistic. There will be issues we need to work through to make sure it all works happily together, but that's the sort of thing you guys pay us for when you buy the game

Steve

[aka_tom_w]

attlefront.com

Administrator

Member # 42

posted January 24, 2005 01:01 AM

Never confuse an implementation of something with what is possible or not. Meaning, just because you play Game X with Feature Y does not mean that Game Z will play exactly the same. If you want a dose of reality in that sense, pick up the other 3D tactical WWII game out there and see how closely it compares to CMBO What I mean by this is that nobody should judge what we can do, or not do, based on what some other group of guys did 10 years ago in a 2D enviroment that worked on a 486 (or a Mac IIci in my case ).

Now, as it so happens you guys that were painfully hoping to rearranging your CC men probably were wasting your time. While CC was certainly 1:1 Graphically, it was not 1:1 Modeled. Unless my memory is playing big tricks on me, much to the contrary of sycophant ramblings a CC unit was no different than a CMx1 unit in that all LOS, LOF, and location calcs were based on a single spot. That means 3 guys of a CC squad on the wrong side of the wall were likely in no danger if the system though of the center of the unit as being on the correct side of the wall. If the system thought the unit was on the incorrect side of the way, the 3 guys on the "protected" side were vulnerable even though they looked protected. I don't remember exactly how the CC system worked, but like I said... unless my memory is getting bad it worked pretty much like this and not like you guys think it did.

Having said that, it isn't as hard to control individual guys within a single unit as it is to control multiple units. The reason is that telling something like a rifle team (to use CC scale) to move to location X effectively defines where the 3, 4, or 6 guys should go to. Now, think of telling a platoon to "take x objective". The AI now has to move several pieces, which are inherently separate from each other, in a way that makes sense. In real life this could mean leaving one Squad (of two teams) behind, moving one Squad (of two teams) down a road 30m to the left, and telling the other Squad (of two teams) to do a zig-zag assault over an open field, but only after the second team arrives on target.

See the difference? If not, read this

The less variation there is in general location, type, and expected behavior, the easier it is for an AI to get it right. The more varried the locations, types, and behaviors the more complicated. Telling the AI to move 6 or 12 guys to "go to this wall and lay down suppressive fire" is a lot easier for the AI than "move these 34 guys in any one of a million ways to take that farmhouse".

Git it now?

Steve

Battlefront.com

Administrator

Member # 42

posted January 26, 2005 03:21 PM

As you guys have all fully grasped, the 1:1 representation is not as simple as blitting a few extra sprites onto the screen. This has been the most frustrating thing to explain to more general gamers. They think that if it appears, visually, in the game then everything is taken care of and we'll all live happily ever after. NOT!

One thing to keep in mind is the general philosophy we use when designing stuff. And that is, the closer we make the basic model relate to the real world, the easier it is to solve problems with making the simulation act realistically.

In CMx2 terms this means simulating each individual soldier and their function within the unit. We need to do this because each unit has some sort of unique mix of tasks for its individual members. For example, a mortar team might have the same number of men as a heavy MG team, but their "jobs" are totally different. A squad with a LMG and 10 riflemen behaves differently than a 6 man recon squad armed with SMGs. So on and so forth.

Once we have the guys individually pegged to different spots, then we can assign more individualized behaviors for the TacAI to use. For example, having a LMG team stay put while the rifle section moves forward on an assault. This is inherently possible because CMx2 already knows what a LMG team is and what a rilfeman is, if for no other reason than to get the animations correct.

Now comes the realism benefit. We have soldiers with individual tasks and individual animations. We also have a range of realistic orders to cover various types of functions. Terrain is also laid out in great detail, as are all the combat modeling that goes along with it (weather, LOS effects, etc.). Weapons and other things are obviously modeled in detail as well. This is a rich pallet of stuff to draw from.

What we need to do is make a TacAI that understands which soldiers should be doing what in one situation vs. another. While this is certainly no small task, it isn't as bad as it might appear to you guys. One reason is that we have all the realism elements to implement standardized military "drills" without much in the way of compromise.

For example, let us say that a Squad has two internal Teams, one of which is a LMG and the other is armed with rifle type weapons. The drill for that nation might state that the LMG team, on the assault, sets up and provides covering fire for the rifle team as it moves to its objective. This is fairly easy to code into the TacAI.

The problems come, as they often do, with fleshing this system out to cover "all" circumstances "all" the time "everytime". That's a challenge. We are confident enough that we can do this. In fact there is only one design for CMx2 and that design involves 1:1 representation. It will happen and we expect the results will be better than average

Steve

[Abbott]

When asked, "What do you want for Father's Day?" Of course I replied "Just you baby" NOT! I said "Get online and pre-order me T-72 Balkans on fire! Woot!

[aka_tom_w]

Posted in the CMx2 casualties thread....

re-post here as this is where some folks may be looking for all the latest news:

Battlefront.com

Administrator

Member # 42

posted June 23, 2005 04:17 PM

This is all very tough stuff for us, as you all (thankfully) seem to understand. Which is why so few games have even superficially dealt with these issues. It's kinda like asking why nobody has brought a flying car to the marketplace... great idea, but there are a "few" practical issues associated with this

As I said earlier, we aren't interested in making a medic/evac simulation, but we are hopefull we can make some improvements to the way things worked in CMx1.

What are we going to be able to do? I honestly don't know. We've pushed all the designs (and there are several) off our plates for now because there are more important things to focus on. After we get a bunch of those things done we'll dust off the designs and see what we have time/energy to do and (most importantly) what will fit in best with the rest of the game.

One thing to remember from earlier conversations... the CMx2 engine will be a work in progress for as long as we are using it before moving onto CMx3. We would have liked CMx1 to have been a work in progress, but unfortunately we found that it was too difficult to work on and therefore little progress was possible without inordinate investments of time and resources.

What this means is that for the first release of CMx2 there will be lots of stuff that can be improved upon for future releases. Some things will never be fully fleshed out (it is impossible to simulate EVERYTHING in minute detail, obviously), but the more important stuff will be improved upon as we move from product to product.

I'm going to guess that WIA/KIA issues will be one of these features that's improved over time. It's something that would be nice to do to the 10th degree in the first game, but based on the utter lack of it in wargames in general, it's clear it isn't something we MUST do for the game to be successful in everybody's eyes.

Steve

[aka_tom_w]

Old bones to chew on

(not new)

Battlefront

Administrator

Member # 35

posted January 03, 2005 04:38 PM

Hi guys,

I'll toss out a few more things for you.

First, the CMx2 engine is not some sort of universal simulator. Each game that comes from it will require coding, fresh designs, customized UI, etc. We are not doing the sort of "change the data and call it a new game" system that so many other wargame developers have done in the past. Instead, the CMx2 engine is more like a developer's toolbox that will allow us to create games quicker and with less reliance upon Charles than CMx1 ever could.

We are still shooting for a release about this time next year with another to follow about 6-8 months later. When we announce the first title we will also announce the second, but nothing beyond that.

While it is true that we don't have any direct competition, we do have competition. We also can not assume that there isn't someone we don't know about looking to copy us. That is one reason we're keeping tight lipped.

The other reason is that it doesn't do anybody any good to talk about stuff that may or may not happen three years from now. That's a long time away and much can change between now and then. All you guys should care about is that in three years time there will be more stuff to play with, and if you aren't crazy about it that something more will be coming soon after.

Oh, and the new graphics capabilities of CMx2 is on par, if not superior to, anything you guys have played thus far or are likely to in the next year. It's amazing what can be done with a fresh slate, radically more powerful hardware, and years of experience

Steve

[ January 03, 2005, 02:35 PM: Message edited by: Battlefront ]

Battlefront

Administrator

Member # 35

posted January 06, 2005 02:30 AM

Some more quick thoughts...

I'm not surprised nobody has risen to the challenge of toppling CM's standing. The teams that have more resources and good talent are run by big companies that have long ago written off our niche. The others tend to be focused on easier things to do. No slight on those guys, just a reflection that we're a hard act to follow.

Kip, there will be a much more involved campaign system. I won't say more than that, but the focus of the game is far more campaign oriented than Quick Battle or stand alone scenario (those options are of course still available). This is part of the evolution aspect of CM. We spent so much time getting the battle stuff right in CMx1 that we had to economize the campaign design. This time 'round we don't have to.

ASHBERY76 , those models are WAY more detailed than they need to be for a game. Having said that, CMx2's models will look far closer to those than to CMx1 models. In theory we COULD put in something that huge, but I don't think anybody would be too happy with the results.

GPIG, hehe... Finding Nemo 3 - Nemo Goes To War! I love it Seriously, the graphics capabilities of CMx2's engine are on a par with the best we've seen from games in development these days. We feel it beats games already released. So yeah, it will be pretty darn good. And when we start to expand our titles post CMx2's first release, perhaps we'll need to talk. I mean, if you aren't too busy making the underware renders for The Incredibles 2, perhaps we can use your services

We are also mindful of hardware demands. Current systems should be able to handle it OK. Stuff from before might have problems. Anything as old as my creaky G4 400GH or a 1GH Pentium will likely be in trouble. The good news is that with Game #2 following Game #1 so quickly one system upgrade should be good for at least 2 games, if not 3, since the core game technology won't be changing within that timeframe.

Steve

Battlefront

Administrator

Member # 35

posted January 06, 2005 04:23 PM

I need to wipe off some of the campaign drool that is hitting the keyboards. CMx2 will not, at least at first, have some sort of "meta-campaign" system. By that I meean something that would allow multiple people to do campaigns and have their results somehow matter. I am also not talking about campaigns where larger issues, which take place outside of the battle, somehow have an affect on the next tactical battle. Maybe someday, but not with the first release and perhaps not even with the second.

Instead the campaign will be somewhere inbetween a meta-campaign and CMx1's Operations. It will also be more focused and far more "story" oriented (i.e. giving meaning to the battles you fight instead of just fighting). More on this at a later date. Just didn't want you guys having your imagining going wild and then being disapointed when we outline how it works.

Yes we can do pre-gunpowder type environments. Might take a little more work in some respects, but in other ways it will be easier (no vehicles ).

Graphics will be as good if not better than the best of the 1st Person Shooters out there. We don't know of any likely wargame that can match even what CMAK looked like One game was mentioned in this thread, but we too aren't sure when (or if) it will be released. It also appears to be less of a wargame, as you guys would define it, and more RTS.

As for WeGo systems... don't forget TacOps Early Grigsby games also had WeGo. But like Jon said, there have been few and far inbetween since then. I can guess why this is for more mass market type wargames, but never understood it for ones which are designed to be realistic.

Steve

Fan Art by Gpig:

(not an employee of BFC)

[ June 24, 2005, 11:34 AM: Message edited by: aka_tom_w ]

[aka_tom_w]

Battlefront.com

Administrator

Member # 42

posted January 24, 2005 01:01 AM

Never confuse an implementation of something with what is possible or not. Meaning, just because you play Game X with Feature Y does not mean that Game Z will play exactly the same. If you want a dose of reality in that sense, pick up the other 3D tactical WWII game out there and see how closely it compares to CMBO What I mean by this is that nobody should judge what we can do, or not do, based on what some other group of guys did 10 years ago in a 2D enviroment that worked on a 486 (or a Mac IIci in my case ).

Now, as it so happens you guys that were painfully hoping to rearranging your CC men probably were wasting your time. While CC was certainly 1:1 Graphically, it was not 1:1 Modeled. Unless my memory is playing big tricks on me, much to the contrary of sycophant ramblings a CC unit was no different than a CMx1 unit in that all LOS, LOF, and location calcs were based on a single spot. That means 3 guys of a CC squad on the wrong side of the wall were likely in no danger if the system though of the center of the unit as being on the correct side of the wall. If the system thought the unit was on the incorrect side of the way, the 3 guys on the "protected" side were vulnerable even though they looked protected. I don't remember exactly how the CC system worked, but like I said... unless my memory is getting bad it worked pretty much like this and not like you guys think it did.

Having said that, it isn't as hard to control individual guys within a single unit as it is to control multiple units. The reason is that telling something like a rifle team (to use CC scale) to move to location X effectively defines where the 3, 4, or 6 guys should go to. Now, think of telling a platoon to "take x objective". The AI now has to move several pieces, which are inherently separate from each other, in a way that makes sense. In real life this could mean leaving one Squad (of two teams) behind, moving one Squad (of two teams) down a road 30m to the left, and telling the other Squad (of two teams) to do a zig-zag assault over an open field, but only after the second team arrives on target.

See the difference? If not, read this

The less variation there is in general location, type, and expected behavior, the easier it is for an AI to get it right. The more varried the locations, types, and behaviors the more complicated. Telling the AI to move 6 or 12 guys to "go to this wall and lay down suppressive fire" is a lot easier for the AI than "move these 34 guys in any one of a million ways to take that farmhouse".

Git it now?

Steve

Battlefront.com

Administrator

Member # 42

posted January 16, 2005 11:40 PM

Tom, I was pointing back to earlier comments I made to reassure people that we wouldn't be yanking away control of units for the simple sake of reduing Borg issues.

Here is an example of how reducing the Borg reduces certainty and introduces natural delays:

Let's say you are on the defensive and suddenly find a section of your line attacked by a bunch of enemy infantry. In CMx1 you could assign all sorts of assets to attack this enemy infantry without significant delays. If you felt the enemy was sufficently tied up you could then rush reserves into the spot to reinforce your defenses. All of this could be done within a turn or two because.

If all we did was simply make each unit capable of targeting only what it spotted on its own, the situation as described would likely play out very differently. Think about all the assumptions and reassurances one has with the current system. Step through the situation in your mind and picture all the things decisions you would make and note how many of them involve KNOWING something that shouldn't be known. Now think about the same situation assuming a more realistic knowledge and reaction system, then see how that might affect your decisions.

I expect you will find that you'd have to hesitate before making plans and then take longer to get plans moving. And if you don't, then think about the possible ramifications if your "rushed" plans fail. Can you afford to take that sort of risk? Sometimes yes, most of the times no.

Put another way... think of how many tried and true "gamey" tactics didn't work in CMBO. Then remember how many from CMBO didn't work with CMBB (oh boy do I remember those discussions ). It is very safe to assume that a good amount of what worked in CMx1 will not work in CMx2 for the same reason (i.e. that we are making the game more realistic).

Battlefront.com

Administrator

Member # 42

posted January 17, 2005 02:51 AM

Remember that 1:1 representation for a battalion level 3D wargame was simply impossible to do. Still is if you're talking about the average computer wargamers have in front of them. We're counting on the average wargamer having to finally bite the bullet and upgrade by the time CMx2's first game comes out. I know I'm long overdue for an upgrade! Cripes, Apple's new $500 buck book sized computer is nearly four times faster than my desktop

At the risk of causing people to ask for a lobotomy sooner rather than later, I've started up a new thread:

http://www.battlefront.com/cgi-bin/bbs/ultimatebb.cgi?ubb=get_topic&f=30&t=003451

I am going to stay away from talking specifics about CMx2's design, but I thought it was time to refocus people on looking at the core problems with simulating command levels before thinking up ways to deal with this age old problem.

Steve

NOT a bone but worth repeating IMHO

Cpl Steiner

Member

Member # 11057

posted January 24, 2005 10:36 AM

"Full Spectrum Warrior" is a good example of the "template" idea mentioned above. When a team is ordered to move, each man's future position is shown graphically (with the AI positioning them close to the best cover) and the player has to confirm the order before they move. This way they move realistically and stack up behind car wrecks or around the corners of buildings. You can even order them to "bound" - i.e. half the team covering whilst the other half moves.

I don't see why this couldn't be implement in CMX2 with a slight reduction in graphical detail (i.e generic terrain rather than pre-designed levels). FSW is much lower level than CM but it does a good job of portraying realistic troop movements.

[ June 24, 2005, 11:28 AM: Message edited by: aka_tom_w ]

[aka_tom_w]

This is NOT a recent bone but it should be re-posted here as well

Battlefront.com

Administrator

Member # 42

posted March 02, 2005 01:22 PM

Wartgamer said:

quote:

Combine this with even normal human intelligence and the AI stands no chance.

end quote

.

Steve says:

This is the crux of the problem. Nobody has made an AI that can think like a Human. Chess AIs are good, but they do not "think" in the way we're talking about in wargaming terms. A good chess AI is nothing more than a very good crunch clearly defined and extremely limited options until it finds the best pattern of them to use. Wargaming AI is, unfortunately, more like trying to simulate a CEO of company and everybody that works for him.

The upshot is that the more the player's actions are dictated by his thought process, the more difficult it is to create an AI that can do anything more than pose a small challenge. The opposite is also true since the less control the player has over the way the battle unfolds, the less the player's superior intellect comes into play. And that is realistic since in the real world no one person has the sort of control over his force that the Human player does in CM.

As for the Scenario Tools being "cup holders"... oh boy is that ever wrong One of the major problems for AI is coming up with a plan based on highly variable situations. CM has more variables than most HUMANS can handle, so to expect the AI to be able to grasp all of these things better than a Human is simply unreasonable. The solution is to allow the Scenario Designer to customize the AI to the scenario and therefore remove some of the need for the AI to "wing it". It is one of the only, and I mean only, ways to improve the AI's performance in a given scenario. Wishful thinkers need not respond

The topic of cheating is a very interesting one that we haven't discussed here since the old days of pre and post release CMBO. The current CMx1 AI does not cheat, and neither will the CMx2 AI. It is the wrong way to go about AI design. However, there is something which on the surface is "cheating" but when looked at philosophically it is more like "simulated intuition".

How does a Human decide what to do in a given situation? Does he simply look at the information in front of him and stop right there? Not a good player, that's for sure. Instead he imagines various possibilities and formulates predictions, which in turn he bases his plans on. The better the imagination and the more insightful the predictions, the better the plans are likely to be. (all else being equal). This sort of thinking is not something we can program into the AI. And I've said this before... if we could do this, we'd stop making wargames and instead make billions producing AI products that nobody else is currently capable of doing

Any ninny can see that this important part of "thinking" is critical to how well an AI can work in a chaotic, unpredictable, and highly variable virtual world. And since we can't recreate this thinking in code, the AI is inherently at a disadvantage so fundamental in nature that it is basically impossible to come up with an AI that will pose more than a passing challenge to the average gamer. That is, of course, if we don't work around this problem.

The work around is to allow the AI to "guess" like a Human would, but without having to write an AI that would win us a Nobel prize. One way to do this is to allow the AI some amount of information that the Human doesn't have in the strictest sense, but DOES have through intuition, experience, or an understanding of the game system's limitations. For example, if you are playing a 1500 point game where the enemy is on the offensive... when you see a company of infantry you can, roughly, figure out how much armor the guy has. The AI, in current CMx1, can not. But if the AI were allowed a certain, highly controlled method of "peeking" at the Human's units... well, that would even things up a bit. The Human's guess based on no peaking will probably be more useful, but it is still something that can boost the AI's decision making capabilities substantially without certain cheating.

By certain cheating I mean if the AI knows you have an unarmored HQ at coordinates 32,45. This is information that the Human player could never have through intuition, since even a good guess is still a guess whereas 32,45 is no guess at all. Therefore, allowing the AI that sort of information is not good at all. It is unrealistic and based on other games which do this VERY frustrating to the Human player. So we aren't talking about this sort of thing.

Instead the AI might know that the player has an unarmored HQ somewhere in the area of 32,45 after having legitimately spotted the rifle companies. This is something that a Human player would likely deduce, then from there check out the terrain and say "hmmm... if I were the other guy, where would I set up my HQ?". The player might then choose a spot and hit it with artillery, or perhaps utilize some sort of recon asset to see if the guess has merit, then act if there is confirmation.

In the end there is only so much we can do. AIs, contrary to hardcore Grog thinking, don't sell games. They don't even hold back sales of games when the AI is discovered to be horrible (like Warcraft II, for instance). But we don't like to avoid trying things simply because everybody else and their brother does. So we'll do the best we can and for sure produce a better AI than other games. But will it be the wet dream of perfection Grogs demand?

No.

That is impossible and therefore we'd be fools to say we can do it.

Steve

The ONE sure thing Steve has avoided any commnent on is asymetrical FOW settings or asyemtrical realism settings to allow a player to give some form or "handicap" or simulated advantage to another weaker human player or the enemy AI. There should be a way in CMX2 to allow the human player to play the game on the most realistic FOW setting (the REAL real hard one like Extra Hard Enhanced Extreme Fog of War with all the hardest inter-unit communication protocols compromising and preventing intel between distant and isolated units, by this I mean the hard from or Relative Spotting setting) and then allow the computer AI to use something like Partial Fog of War or some limited "simulated intutition" or "hinted enhanced" computer Articial inteligence to play the game against, thus providing the player with a NON-cheating but decidedly "articifically enhanced" enemy AI to provide perhaps a slightly more challenging gaming experience.

Combine the possibility of asymetrical FOW settings with the possibility that scenario designers will be able to "program the AI" or somehow script AI or configure SOP settings for the AI player in a specialized scenario's designed to be played by human players against a specific set-up of enemy AL controled units (ESPECIALLY on defense) and you will see many new players to CMx2 complain that some scenario's are unwinable or too hard to "beat".

I for one hope the CMx2 game disk is released with at least one or two CMx2 scenarios that are so tightly organized or well scripted by the designer as to be nearly impossible (in a historically accurate way and setting of course) for the human player to have any chance of a victorious result no matter how well they play or how lucky they get. (not all scenario's on the disk just a couple of real good juicy challenging ones that are near impossible to win for the human player against a well scripted AI opponent and by that I am NOT looking for replayability I am just suggesting that with FULL FOW and playing the scenario BLIND for the FIRST time, the human player, no matter how skilled or how lucky will get smoked by the well scripted AI opponent, a scenario like that as a DEMO scenario for the release of the CMx2 engine would be greatly appreciated as well HINT HINT! )

But now I guess I am just being an optimist....

-tom w

[ July 01, 2005, 08:44 PM: Message edited by: aka_tom_w ]

[aka_tom_w]

The ONE sure thing Steve has avoided any commnent on is asymetrical FOW settings or asyemtrical realism settings to allow a player to give some form or "handicap" or simulated advantage to another weaker human player or the enemy AI. There should be a way in CMX2 to allow the human player to play the game on the most realistic FOW setting (the REAL real hard one like Extra Hard Enhanced Extreme Fog of War with all the hardest inter-unit communication protocols compromising and preventing intel between distant and isolated units, by this I mean the hard form of Relative Spotting setting) and then allow the computer AI to use something like Partial Fog of War or some limited "simulated intutition" or "hinted enhanced" computer or simulated "situational awarness" (like I have ANY idea what that means Doh! ) Artificial inteligence to play the game against, thus providing the player with a NON-cheating but decidedly "articifically enhanced" enemy AI to provide perhaps a slightly more challenging gaming experience.

Combine the possibility of asymetrical FOW settings with the possibility that scenario designers will be able to "program the AI" or somehow script AI or configure SOP settings for the AI player in a specialized scenario's designed to be played by human players against a specific set-up of enemy AL controled units (ESPECIALLY on defense) and you will see many new players to CMx2 complain that some scenario's are un-winable or too hard to "beat". (please !!!)

I for one, hope the CMx2 game disk is released with at least one or two CMx2 scenarios that are so tightly organized or well scripted by the scneario designer as to be nearly impossible (in a historically accurate way and setting of course) for the human player to have any chance of a victorious result no matter how well they play or how lucky they get.

(Not all scenario's on the disk, just a couple of real good juicy challenging ones that are near impossible to win for the human player against a well scripted AI opponent and by that I am NOT looking for replayability I am just suggesting that with FULL FOW on, AND playing the scenario BLIND for the FIRST time, the human player, no matter how skilled or how lucky will get smoked by the well scripted AI opponent, a scenario like that as a DEMO scenario for the release of the CMx2 engine would be greatly appreciated as well HINT HINT! )

But now I guess I am just being an optimist....

-tom w

[ July 13, 2005, 07:58 PM: Message edited by: aka_tom_w ]

[Gpig]

Just in case you missed this wee bone, Tom:

Recent post by Steve of Battlefront, in the Wartgamer/Rexman "HE Effectiveness" thread.

The problem in CMx1 is that original coding did not allow small arms weapons and mortars to be distinct from crews. They are one and the same. It wasn't the best design decision we ever made, but it was a very early one that couldn't be undone without redoing core parts of the game system. So it remained in all three titles.

Abandoning crew served weapons or vehicles was deemed to be a tough thing for us to simualte without opening things up for abuse. With about 100000 other things not coded at the time, we decided to just go with the abstraction you guys are familiar with and move on to other things. A good decision at the time, but unfortunately it was not one we could undo later on.

CMx2 fixes both of these issues. Soldiers and weapons/vehicles (of all types) are not welded together. They are separate.

Steve

[hellfish]

Quick hypothetical question for Steve:

Would the CMX2 engine be able to handle Grand Tactical battles? For example, instead of handling individual tanks and squads, you'd maneuver platoons as single units. Basically something like at 1/10th the scale of the game now, where a terrain cell would be 100m instead of 10m.

Just asking.

[Battlefront.com]

fytinghellfish,

Yes, in theory the engine could be adpoted to do more Operational and Strategic type games. It wouldn't be a quick conversion since the game system would need massive changes, but it could be done. We currently have no plans to do this, but it is something we've thought about.

Steve

[hellfish]

Fab. Thanks.

[aka_tom_w]

Thanks Gpig

sorry I missed that one....

if anyone else here finds any new juicy bones buried elsewhere and wants to add them to ths thread PLEASE feel free to go right ahead just like Gpig did.

-tom w

Originally posted by Gpig:

Just in case you missed this wee bone, Tom:

Recent post by Steve of Battlefront, in the Wartgamer/Rexman "HE Effectiveness" thread.

</font><blockquote>quote:</font><hr /> The problem in CMx1 is that original coding did not allow small arms weapons and mortars to be distinct from crews. They are one and the same. It wasn't the best design decision we ever made, but it was a very early one that couldn't be undone without redoing core parts of the game system. So it remained in all three titles.

Abandoning crew served weapons or vehicles was deemed to be a tough thing for us to simualte without opening things up for abuse. With about 100000 other things not coded at the time, we decided to just go with the abstraction you guys are familiar with and move on to other things. A good decision at the time, but unfortunately it was not one we could undo later on.

CMx2 fixes both of these issues. Soldiers and weapons/vehicles (of all types) are not welded together. They are separate.

Steve

</font>

[jim crowley]

posted July 15, 2005 02:45 AM

--------------------------------------------------------------------------------

We are many months away from having anything like a Demo to show you guys. I mean, even SCREENSHOTS of a Demo like environment. Unlike CMx1's development, the models and graphics are being implemented far more towards the end. For now we are using a very basic set just to test the engine's functionality. That means not so much to show until we get much closer to the game being done.

I know some of you guys will look at the above and think this is somehow a BAD thing. But really, it is the best thing possible. The whole point of CMx2's engine is to make adding models, textures, sounds, and soforth relatively easy. Therefore, instead of needing 1-2 years to flesh out a game's graphics we will only need a few months. Not so good for advanced screenshots, but a heck of a lot better in terms of getting the next game out quicker.

Don't worry... we'll show you something fairly soon. Just don't expect to see mouse holing and tons of varried terrain and models. That's a ways away still.

Steve

Hmmm. Finished by the end of 2005? Doesn't seem very likely.