Length of game.

[wodin]

It seems to me at 26 turns quite a quick game for 4 years of fighting.

How long would one game start to finsh take on average?

I would have thought a turn = month would have been just right.

[DoomedMantis]

I would agree, a lot can happen in a month let alone 2. How does weather get modeled for this for example? Could have one week of perfect weather and the rest rain, but organise an attack based on the good weather, and run into disaster.

[*Lava*]

Hi!

A lot happens in a game turn.

You start with what is called the "interphase" portion of the turn. Here, all your "admin" stuff happens: orders for fleets, R&D, diplomacy, air deployments to various fronts, unit builds/reinforcements, draft call-ups, etc. After this phase, you move to the Naval Phase (where fleets carry out orders) and then the Operational Phase. The Operational Phase is where you determine whether or not you will "activate" army HQ's and attack. From here you move to the Orders Phase. This is where Armies and their Corps are given movement orders and artillery are ordered to conduct bombardments. When both sides have completed the Orders Phase the computer conducts the Resolution Phase. Since this is a WEGO system, bombardments from both sides are carried out first. This is then followed by simultaneous movement and any battles which may occur are resolved. After the Resolution phase, you then return to the Operational Phase and go again (note, you only have 1 Naval phase per turn). Each cycle, (Operational, Orders and Resolution) is referred to as in "impulse." Most turns will have 2 impulses, approximating 2 months of the war. August 1914, for example, has 4. When you have completed all the impulses for eact turn, you move to the next Interphase or turn if you will.

In outline form the August 1914 "turn" looks like this:

August 1914
-- Power Interphase
-----Impulse 1 (Naval, Operational, Orders, Resolution)
-----Impulse 2 (Operational, Orders, Resolution)
-----Impulse 3 (Operational, Orders, Resolution)
-----Impulse 4 (Operational, Orders, Resolution)

So as you can see, "turns" are actually quite lengthy. What's going on here is this, as I understand it. A "turn" represents an operational planning cycle, while the "impulses" represent the actual implementation of that cycle. So for example, in August 1914, with 4 impulses, you have the the opportunity to attack 4 times. As the Germans, you may choose to push into Belguim and begin your swing into France. However, as you push your troops, they will recive losses in battle in both strength and readiness. The harder you push, the faster the readiness drops and with it your troops ability to conduct "effective" attacks.

As for weather, that I am not really sure about. I belive it is an abstraction used in the calculation of such things as readiness, depending on the time of the year.

Hope that sheds some light on "turns" for you guys. From the testing side, "balancing" is still going on in areas such as combat, artillery effectiveness and readiness.

Ray (alias Lava)

[DoomedMantis]

I understand that the turns are lengthy, but I also understand that 2 months is a hell of a long time. I will be interested to find out what the testers have to say on this.

[*Lava*]

Hi!

ORIGINAL: DoomedMantis

I understand that the turns are lengthy, but I also understand that 2 months is a hell of a long time. I will be interested to find out what the testers have to say on this.

I am doing pre-beta testing. This game is not Strategic Command set in WWI. You have to remember this is an operational game on a grand scale. It works real well. The feel is nice. When you attack with the 1st German Army what you get is not a battle as in "The Battle of Waterloo," you get a battle as in "The Battle of the Marne." Such an encounter is of enormous proportions, and represents not just the battle itself, but all the planning, logistics, and maneuvering that goes with it.

As I said, a 2 month "turn" is normally represented by 2 impulses, which basically gives you an impulse per month, expect in the summer when there are more. The scale works well, in my opinion.

Ray (alias Lava)

[wodin]

So in effect there are four turns within a turn?

This sounds a lot better than what I originaly thought.

Thankyou for the prompt reply.

Good to see its a wego system aswell, this gets better all the time.

[*Lava*]

Hi!

ORIGINAL: wodin

So in effect there are four turns within a turn?

This sounds a lot better than what I originaly thought.

Thankyou for the prompt reply.

Yes, for August of 1914. Most turns are actually 2 month affairs with 2 impulses. After 1914, the summer months get 3 impulses in the Great War scenario.

Ray (alias Lava)

[DoomedMantis]

Good to know, thanks for the info

[wodin]

The whole system for this game sounds very good indeed.

Implementing a lot of things Ive wanted to see in other wargames.

The fact that it is WW1 is a joy and I hope it sells well as some seem to dismiss this era far to quickly.

What level would you class this game in easy/moderate/hard?

Im hoping it is a game which seems straight forward on the surface but is hard to master and suprises you as you play and learn new aspects of the game.

[*Lava*]

Hi!

ORIGINAL: wodin

The whole system for this game sounds very good indeed.

Implementing a lot of things Ive wanted to see in other wargames.

The fact that it is WW1 is a joy and I hope it sells well as some seem to dismiss this era far to quickly.

What level would you class this game in easy/moderate/hard?

Im hoping it is a game which seems straight forward on the surface but is hard to master and suprises you as you play and learn new aspects of the game.

On the single player side, it's hard to know right now because the AI is not ready. Frank has an interesting concept he is working on where the AI will actually "learn" from its human opponent. Too early to tell.

On the PBEM side, I think this game is going to be really open ended. There are loads of different strategies that can be used, and heavy FOW means you can count on surprises.

On the surface the game is very straight forward, the tough part is making it one which continually surprises. There are some kinks, but I have a pretty strong conviction that this game is going to be fun. Kinda like the old days.. if you know what I mean.

Ray (alias Lava)

[DoomedMantis]

thats fantastic news, hope he pulls it off

[FrankHunter]

Just to add to what Ray has already said.

There are 6 turns per year made up of from 2 to 4 impulses depending on the season.

If you play from 1914 to 1918 it can be a pretty long game but there are different start and end points. 1914 should be playable within 1-2 hours depending on your play type.

[Fred98]

Some of the descriptions above, concerning logistics especially, seem to imply, that this game could be described as: “Uncommon Valour in the Trenches”

If it is then I am in!

Perhaps you could explain further?

[JamesM]

Interesting point about Lava about the leaning AI. The Road to Moscow touted that concept. I do not how far they got (that programs appears to have become vapourware).

Jim

[*Lava*]

Hi!

While I can't speek for Frank (who knows the best about this), what I understand the plan is, is to have game testers play the game and then send it to Frank, who will then incorporate the game testers game experiences (strategy's, battle tactics, etc.) into the AI. So the game will start with some "experience."

I'll give you an example. If I (as the German player) use on a fairly frequent basis a strategy of bombarding an enemy hex before assaulting it, the AI will learn this and will give it back to me when I play the French. If you understand what I mean..

Add to this a random AI "personality/strategy" and hopefully what you get is a game that doesn't play the same twice.

Seems fairly innovative to me. Hopefully, he can pull it off.

Due note, this is my "impression," Frank is the one who really knows what's going on.

Ray (alias Lava)

[Hanal]

ORIGINAL: Lava

Hi!

While I can't speek for Frank (who knows the best about this), what I understand the plan is, is to have game testers play the game and then send it to Frank, who will then incorporate the game testers game experiences (strategy's, battle tactics, etc.) into the AI. So the game will start with some "experience."

I'll give you an example. If I (as the German player) use on a fairly frequent basis a strategy of bombarding an enemy hex before assaulting it, the AI will learn this and will give it back to me when I play the French. If you understand what I mean..

Add to this a random AI "personality/strategy" and hopefully what you get is a game that doesn't play the same twice.

Seems fairly innovative to me. Hopefully, he can pull it off.

Due note, this is my "impression," Frank is the one who really knows what's going on.

Ray (alias Lava)

If this is the case then it seems that there might be an advanced and detailed scripting element in the AI which would make it a very challenging game....

[Dirtdog20]

Now that Frank is back maybe he could join in the discussion on this if he wants to.

[FrankHunter]

Sure, the AI I planned has been implemented, it tracks what it and human players do and records a few hundred past games. But I'm still not happy with it, however, the parts I'm not happy with could be due to the fact it doesn't have enough information yet and much of what it has comes from playing itself. I'm sure its just a few kinks that need to be followed up but we'll see.

[pzgndr]

the AI I planned has been implemented, it tracks what it and human players do and records a few hundred past games

Is this going to be a game feature or option, or just something you're using for the AI development? Reason I ask is that I wouldn't want "my" game AI being recalibrated based on any mistakes I make while learning how to play the game.

Now, later on after I get better and start anticipating AI behavior, then it would be nice to toggle on the AI "learning mode" or whatever you call it. [8D]

[FrankHunter]

The AI is not an option, so it can't be turned off. However, when the game is shipped it will already contain a library of hundreds of games. The idea is that if you have a "best plan" the AI will learn that and hopefully be able to counter it. So Germany sending the same 30 corps through Belgium may encounter little opposition the first time its done but by the 2nd or third time there should be 30 French corps ready to meet it. Also, the AI is likely to try setups and plans against you that you used against it.

The AI for naval and R&D, diplomacy etc works the same way. It learns from what players have done.

There is also a hard-coded AI that acts as the fall-back if there are no prior examples of a particular situation.