Prepare yourself for a wargaming tour-de-force! Conquest of the Aegean is the next generation of the award-winning and revolutionary Airborne Assault series and it takes brigade to corps-level warfare to a whole new level. Realism and accuracy are the watchwords as this pausable continuous time design allows you to command at any echelon, with smart AI subordinates and an incredibly challenging AI.
HttR is revolutionary. But IMHO, is not the continuous space and time that makes it so. Panther created a command structure and then a wargame around it.
ORIGINAL: Michael K Lauren and Roger T Stephen. "Using fractals to understand the role of entropy in complexity science: an examination of the MANA combat model". Article published in the journal Fractals
This leads us to construct an alternative view of warfare. The most potent force is the one with the best command structure. The objective of war is to create enough disorder to unravel the opposition’s command structure. Weapons are one instrument for doing this. A laterally minded commander might well employ several other methods, such as maneuver and shock. This offers an explanation as to why apparently very similar armies perform dramatically differently on the battlefield. Perhaps a historical example of this is the defeat of Darius III by Alexander the Great. Both sides were armed with similar weaponry, and Alexander’s forces were significantly outnumbered. Yet Darius’ forces fell to pieces in the heat of battle. This is strongly suggestive of a lack of cohesion in Darius’ army, so that it was vulnerable to the creation of disorder within its ranks. Although Darius’ army no doubt had a hierarchy, the question is whether Darius’ officers were competent enough to prevent disorder spreading up the command chain.
ORIGINAL: hank
The main point was being able to control the units (whether battalion or regimental, etc.) in the way they actually were. i.e., line them up and tell the to march to battle ... shoulder to shoulder ...
hank,
did you try Sid Meier's Civil War games ? You direct regiments, composed of several batallions. If you give order to the regiment commander, he directs his men to battle it the formaion of your choice : line, double line, skirmish, column. You can give orders at batallion level if you so choose. There are Napolonic incarnation as well : Waterloo and Austerlitz.
As far as I now, the MadMinutes Civil War game does that as well. With more style and updated graphics. But I never played it. It should change soon.[;)]
What you all are chatting about here is what I call the "third tier problem."
Having been a beta tester for MMG's Bull Run game I can say that their friendly AI is indeed quite good, but still it has not been able to break the code on the "third tier" of command.
Games have been able to implement a two tier command structure for quite some time. That is you give a command to a brigade, for example, and the brigadier orders his regiments to properly position themselves. I first saw this in Frank Hunter's ACW game. Here you had an army commander who you could give orders too and the commander, based on his personality, would order subordinate units to execute those orders.
In MMG's game, you start with the individual soldier. Each sprite actually has its' own little AI. The little dude will maintain unit integrity and decide on what target to shoot at. This is done automatically by the AI, it is a script. The soldier is part of a regiment. The regiment can be commanded directly by the player or by the regiment's brigadier. This is your classic 2 tier command system. The interesting thing here is a twist has been added, stances (or basic standing orders, if you will). If you give the brigadier an "attack" stance, for example, and then send him through a wood where he can "see" the enemy, he will then begin ordering his brigade to attack that enemy. Very kewl. However, although I am not a programmer, I do not believe this is actually a "real" second tier. I believe the first tier (the regiments) have a scripted response to a general situation... defense or attack as an example. So actually, although it appears like a two tier system, it really isn't, if you get my drift. (If you are confused, don't worry, so am I.) [:)]
Above the brigade commander, you have the division commander, and this is the real second tier in MMG's game. Once again, you can move the division commander to a point where he "sees" the enemy and he, like the brigade commander, will then send orders to his brigade commanders to implement the stance you have given him. Thus, merely moving the division commander to a point where he makes contact, will set off a chain of events which sees brigades and regiments moving into combat and fighting the battle. All this, simply by moving the division commander and nothing more.
The problem is implementing the third tier. In MMG's game this is the army commander/Corps commander. The player assumes this role. The "third tier problem" has been around for quite some time and as far as I know, nobody has as yet cracked it. The reason for this is because, it is at this point where operational "art" really starts kicking in and it is very difficult to script that kind of thing for an AI. Here, it is just not the tactics, but also things such as timing and abstract maneuvering such as feints, envelopements, reinforcement, etc., that in a fluid situation is very difficult to model.
I do, however, think that Norb and Adam over at MMG may very well solve the third tier problem through the use of "personalities" which can help narrow down abstract modeling and lend itself to scripting.
Nevertheless, it is a difficult problem which I have once heard referred to as "the holy grail" of wargaming.
MMG's Bull Run, btw, is really two games. If you are into real micromanagement, you can control ALL the units (regiments, guns) in the game. This is best played using small amounts of forces but almost impossible using the entire OOB included in the game. However, if you WANT, you can play it from an operational stand point and maneuver divisions in the open play option of the game and let the AI take care of messing about with the regiments and brigades. It is quite kewl, and if you are into operational type games, well worth it.
I am looking forwards to the CWBR2. In particular, the next generation of OP. I found CWBR's OP just too open if you know what I mean.
In terms of UI and the command structure, I am still not sure which I like better - PG's control anywhere in the chain of command or MMG's you assume one particular role.
With PG's, the player is less likely to feel peripheral or just a spectator (CWBR - doesn't happen in the designed scenarios, but can happen easily in OP). *
With MMG's, playing only one role does make for some interesting possibilities and forces the player to be more creative in how they contribute to the overall battle.
* NOTE: PG'S - for MP (with more than two players) as a future feature in the series it has been discussed to have players assume different roles within the OOB.
2021 - Resigned in writing as a 20+ year Matrix Beta and never looked back ...
ORIGINAL: Lava
The problem is implementing the third tier. In MMG's game this is the army commander/Corps commander. The player assumes this role. The "third tier problem" has been around for quite some time and as far as I know, nobody has as yet cracked it. The reason for this is because, it is at this point where operational "art" really starts kicking in and it is very difficult to script that kind of thing for an AI. Here, it is just not the tactics, but also things such as timing and abstract maneuvering such as feints, envelopements, reinforcement, etc., that in a fluid situation is very difficult to model.
CWBR! What freakingly good game! Pity my PC is not good enough to run it.
One of the Chapters of the book: AI Game Programming Wisdom 2
is devoted to that exact problem.
The authors claim that commanders at different levels in the hierarchy should have different representations of the battlefield. For example the representation of the battlefield for a Rgt. Commander should be way more detailed (and smaller in reach) than the representation of the battlefield for a Div. Commander. The Div. Commander's view of the battlefield should be way more coarse than the view of the Rgt. Commander. Well, that's how it works in real life anyway.
BTW, the chapter is writen by guys from Atomic Games. Their Close Combat AI sucks big time, so I don't know how successful the approach actually is.
Woody Allen - "Those that can, do. Those that can't, teach."
Got a link to the book on Amazon? I have always been interested in game AI design.
The CC series had one very nice and intuitive interface. Given the CPU cycles now available just think what could be done for those games with just an under the hood AI rewrite. Oh, well ...
2021 - Resigned in writing as a 20+ year Matrix Beta and never looked back ...
Just click on the title in my previous post.
I need to correct myself in something (just pulled the book out of the shelf) . The author is Tom Kent who publishes under the affiliation of Freedom Games (is that Eric Young's previous company, now merged with DAS entertainment?).
Here is the abstract of the book chapter:
Abstract: RTS games tend to handle soldier AIs individually, giving each unit specific tasks from the computer player. Creating complicated, cooperative tactics are impossible for such systems without an immense effort in coding. To develop complex, large-scale plans, a mechanism is needed to reduce the planning devoted to the individual units. Some games already collect individual soldiers into squads. This reduces the planning necessary by a factor of ten, as one hundred soldiers can be collected into ten squads. However, this concept can be taken farther, with squads collected into platoons, platoons into companies, and so on. The versatility such groupings give an AI system are immense. This article will explore the implementation of a multi-tiered AI system in RTS-type games, including the various AI tiers, a set of related maps used by the AI tiers and an example to illustrate the system.
Is a good article. However if this guy wrote the AI for Eric Young's Squad Assault and for the Close Combat series ...
Edit - It supposedly went gold back in February. No "News" since May. Then again if BlackBean is the European publisher, maybe Martix will be the US publisher...
We don't stop playing because we grow old, we grow old because we stop playing. - George Bernard Shaw
WitE alpha/beta tester
Sanctus Reach beta tester
Desert War 1940-42 beta tester
Our Airborne Assault engine is multi-tiered. You can have as many as you like. We currently support everything from Platoon to Army Group. It's the AI which makes the difference. While we apply different values to the different levels - eg orders delay, reaction times, reasessment times, command range etc - essentially it matters little what level of command you are. The command system is very scaleable. So if that's the holy grail of wargaming, then look no further.
But the funny thing in life is that once you have climbed to the summit of one seemingly unclimbable mountain, another even bigger one confronts you.[:)]
Actually, almost from the moment you entered the MMG forums I figured you were doing some work on another project. I even made a post to that effect in our private forums. [:)]
You have the advantage here as I have unfortunately not had the opportunity to try out your engine. Perhaps, that needs to change.
In some ways, from what I can understand by the posts, your engine appears more advanced than MMG's, but the direction of both seems to be bascially along the same track.
Our Airborne Assault engine is multi-tiered. You can have as many as you like. We currently support everything from Platoon to Army Group. It's the AI which makes the difference. While we apply different values to the different levels - eg orders delay, reaction times, reasessment times, command range etc - essentially it matters little what level of command you are. The command system is very scaleable. So if that's the holy grail of wargaming, then look no further.
But the funny thing in life is that once you have climbed to the summit of one seemingly unclimbable mountain, another even bigger one confronts you.[:)]
Intriguing.
I have always leaned towards the command/operational side of gaming rather than the nitty gritty tactical side. Not that I don't want the tactical side presented, but that I like to see what happens after I have planned an op and begun executing it. Most strategic/operational wargames tend to abstract the tactical side, but I would perfer everthing under one roof.
I have a trip coming up soon, but afterwards, I will check out how you climbed the "multi-tiered" mountain. You have definately tweaked my curiousity.
Your COTA AAR would suggest only a two tier chain of command (Divisional Commander with subordinate brigades and support units). Can I imply from your post that this is a "simple" representation of your game, one which does not illustrate the "multiple" tiers which the game can manage?
Both the HTTR and COTA AARs were relatively small scale both in terms of duration (2 days or less) and number of units involved (around 50) on one side. Both games come with much larger scenarios that go from 8-12 days with with hundreds of units. When I say units, I mean counters here with the smallest in scale being companies, support weapons platoons, or tank squadrons. The highest level HQ on the map would be the corp.
At the moment, the only thing limiting scenarios being even of larger scope are:
(1) CPU cycles ... as you know this results in many more calculations for pathing, LOS, and combat resolution as map size increases and unit counts go up.
(2) The UI might require some more filtering tools and OOB improvements to further reduce clutter. HTTR had added friendly unit filters and a small counter view and enemy intel filter. COTA further refined the filters and added an OOB display along with an ability filter and manage global messages.
---
COTA took a further big step towards a future day of much larger scenarios by adding a realistic supply system. Less of an issue for small scenarios, but as they get larger, realism would be further reduced without supply routes and movement of supply via the road network.
---
So, the system is inherently and highly scalable both from a UI point of view and a game mechanics point of view.
(1) It does not suffer from being unable to scale up like the Combat Mission series due to exponential work load on the player as unit counts go up. With HTTR/COTA, the work load for the player increases gradually despite vastly increasing unit counts.
(2) It does not suffer from having to abstracting low level combat and leave the details to imagination like in Victoria, HOI, and HOI2. You can command at the high level, but can follow the battle at a much lower level of detail; similiar to CWBR.
(3) It does not require a two level game system coupling two disparate levels such as the Total War series, Close Combat 4/5, and the upcoming Combat Mission Campaigns.
(4) Although sometimes, I have to admit wishing I could watch some of this in 3D. 2D and counters is much easier to scale and one could argue more realistic when one contemplates command at higher levels.
---
Back to my AARs ... Well, given how much work it takes to compile notes, do screenshots, and write them, that accounts for why I pick the smallest scenarios to use. Otherwise, everyone would be playing the game before I even completed my AAR! Also, I like small scenarios, they are very easy to illustrate the games flexibility in terms of allowing micro/macro management side by side. Finally, small scenarios are better for writing tutorials. Another one of the testers, Gary, likes to do AARs, but he generally prefers large scenarios for his AARs. His AARs tend to focus more on the battle itself and less on the game play.
2021 - Resigned in writing as a 20+ year Matrix Beta and never looked back ...
ORIGINAL: Lava
I have always leaned towards the command/operational side of gaming rather than the nitty gritty tactical side. Not that I don't want the tactical side presented, but that I like to see what happens after I have planned an op and begun executing it. Most strategic/operational wargames tend to abstract the tactical side, but I would perfer everthing under one roof.
This is a game that definitively you need to take a look at.
I won't spoil it to you. But you will see how much fun it is to sit as a Corps Commander in HttR and plan your operations, with your mind on what could happen two hours ahead in the future. It will blow your mind away.
Your COTA AAR would suggest only a two tier chain of command (Divisional Commander with subordinate brigades and support units). Can I imply from your post that this is a "simple" representation of your game, one which does not illustrate the "multiple" tiers which the game can manage?
CoTA, as HttR, has a multi-tiered chain of command (Corps, Division, Brigade, Regiment, Battalion, Company and Platoons, the last generally attached to a Co.).
You can take command at any level. Example: you start an scenario in which a Div.is available and you have to take objective X. You can just click on the Div. Hq. counter and order it to "attack this position". You can select a speed, a route, or later modify the assembly point. Then the friendly AI will figure out how to accomplish that attack. You can just sit and watch the action unfold. Let's suppose that 1 hour and a half later you notice an inviting gap in the enemy defenses, or one of your flanks threatened. Then you can click on whatever Battalion of your choosing (choose wisely, order delays apply) and order them to exploit the gap or protect the flanks. The Battalion HQ you have clicked on will receive the orders and figure out how to move all its Companies and support platoons to achieve the mission you ordered them. The chosen Battalion now becames "detached", that is will not follow the orders that the Divisional HQ has given them at the begining of the battle, and instead will follow the new orders you gave them. Eventually you could start giving orders to the Companies that compose the Battalion, detaching them from the Battalion HQ. It's up to you. But beware on how you manage your command structure, because too much micromanagement eventually hits you back. That is, your OODA loops (observation, orientation, decision and action loops; you can read about them in the strategy guide) will become very long and defeat will be almost guaranteed. That's what I meant in a previous post: command structure is crucial, in the sense that you have to manage smartly how orders go down the chain of command. You can have a lavish SS Division on your hands, but if you don't manage your chain of command accordingly, they will behave like a mob with cool weapons.
Even though HttR depicts a battle that probably suits lots of people's interests more than CotA, no one can minimize the huge improvements Arjuna and Co. has made to the game. So lookout for CotA too.
I own HTTR as well as CWBR and both are fine games at what they do. However, to get back to the original post, I don't believe either will spawn a "new direction" in computer wargames. I don't mean to say that the real time engine will not be used again, it's just that I've always challenged the assertion that turn- and/or hex-based games are "old" and merely a mindless continuation of clunky board game concepts.
Good abstraction is more effective than abstract precision. Real time "realists" claim that war happens continuously without pauses, so that real time games are inherently more "realistic". That's only true if one's design intent is to present the wargamer with as much raw data and confusion as war produces. Many wargames, however, intend to decipher war and present the issues to the gamer in as coherent a manner as possible so the gamer can focus on more of the "fun stuff" and on less of the collateral complexities of war. Thus, the abstractions of the "unit counter", the "hex" and the "turn" become organizers for the human mind to aid in promoting a clean, abstract, but realistic system which can be gamed more certainly than by using nondilineated methods.
The first board wargames did not employ hexes, or squares, or any other kinds of "organizing systems". Kriegspiel was played on contour maps with metal unit pieces scaled to fit the frontage of the unit formation they depicted. Hexes were actually innovations designed to make gaming more user friendly. Since even hexless real time games must have pixel "ranges" at which units are supposedly in contact, it's not really even that great a difference between hexes and hexless terrain. There's always a grid; it's just about how small it is. Instead, the issue is the gameplay involved and the validity of the abstractions employed. A game that purports to model events to an extremely high level of precision risks being more distorted than a game with a larger scale, but better abstraction. The more measurable data samples you can incorporate into your combat system over a broader span of time, the more accuracy you can expect. Unless detail is "fun" in and of itself, the increased detail may even make your game less "realistic" in terms of accuracy.
As for turns vs. no turns, it is the same as reading text with and without paragraphs. Paragraphs organize thoughts and make it possible to read much more text at once than one can read without them. Wargamers can mentally manage a far greater number of units when play is divided into turns than when one must manage everything at once. The turn also allows data compiled from actual events (historical battlefield casualties, e.g.) to be applied more systematically by utilizing set time periods. This also enhances human comprehension of the results of those game mechanics.
I'd note that one of the main weaknesses of hexagons lies mainly in the false normalization of unit frontages to the scale of the hex, especially where in many tactical games unit frontage is paramount. This would be an area where I'd think turn-based hex gaming could use some real innovation. And it's also not clear that nonhex-based games actually handle frontage issues any more realistically -- they should be capable of it, but it needs to be coded properly, like all software.
So I suppose that different methods of presentation are valid and enjoyable depending on the motivation for playing the game and the purpose of the game itself. As far as comparative study and analysis of warfare, turn-based hex wargames have provided me far more insight than any other form of presentation. Personally I believe there's a reason for that. However, as I noted, other presentations have their utility and allure as well, I'm just defending the much maligned "old style" wargames... [:D]
