Turn time

Started by Andy ONeill, June 11, 2020, 06:18:31 AM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

Andy ONeill

Ezrra and I have been discussing turn time.
My thinking is that rather than one variable turn time we should use two.

Why do I think that?
You might ask.

At the end of each turn both players give orders to their pieces.
Due to fog of war, they can't see where all enemy pieces are.
A player is quite likely to give an order to a piece that logically makes no sense as it is executed.
Let's consider a simple example.



Here, red and blue are both in woods and not visible to the opponent.
The Red player is not aware there is a blue piece in that woods ahead and vice versa.

Both players give orders for their pieces to move straight forward.
If they follow their orders then they just move past each other.

Clearly, if these were two battalions then we would expect the colonel of each to spot an enemy unit and give orders immediately for their unit to react.
They don't send an ADC off for orders from on high.
That would be a stupid thing for them to do.
And in fact the expectation was that an officer would act independently.
To the point of completely ignoring a superior's orders.
This was expected because before radio a superior was very likely unaware of considerations obvious to those "on the ground".

In reality.
The Colonel looks at what's going on and makes a decision.
This is going to be influenced by his mission, relative strengths of his unit compared to the enemy unit and so on.
But essentially, the unit commander will take a decision and implement that immediately.

In order to model this and avoid "stupid" things happening I'm going to create something called a decision tree.
This is something like a workflow where there's a decision or check of a condition and a branch to another decision, and so on until a final leaf of the tree what the piece should do.

We have something called Posture which defines the piece's mission in simple terms.
They are attacking, holding, demonstrating ( or others ).

If we grossly over simplify what a decision tree might sort of look like.
Red might look at blue and compare unit power.
Is Red less than half blue?
Yes = retreat
No = check range
    Is Blue within effective shooting range?
       No = consider moving
       Yes = Fire

But these decisions will be automated.
The player will see results but has no input.
No input is a fairly realistic model of what a general would experience.
Once they committed units to battle that was pretty much all their input done.
They could reinforce with more reserves  ( un committed asset(s)).
But beyond that they were pretty much down to waiting until they heard who won.
But no input isn't so popular.

I'm a long term wargamer.
I know players hate loss of control.
Hence it is likely that they will choose shorter turn times to reduce the effect of these automated decisions and increase their opportunities for input.
Gaining more control is not the only effect that will have though.
Initial stages of a battle are usually movement to contact ( usually called the approach ) will take more turns and it'll probably be tedious.

A shorter turn time also suits combat better.
If you choose 20 minutes that is longer than the time some entire battles took to decide.
Once the two sides had closed.
The Plains of Abraham reportedly took 15 minutes between the French getting into musket range to losing the battle and the remnants withdrawing.

Longer turn times to approach also make ambush scenarios more practical. 
When the unexpected happens the initial response should be confusion.
It also discourages unreasonable micro management prior to anything substantial happening.

Thinking about this reminded me of mechanisms in many tabletop skirmish games.
There is frequently a difference between moving to contact and once contact is made.
Contact being once the shooting starts.

First shot is very easily tracked.
We would probably want to add once any piece gets within maximum range of enemy.

With this mechanism you could then have a (say) a 30 minute turn time for approach.
Then (say) 2 minutes once in contact.

This would be fairly easy to implement and all the effects I can think of seem positive.
What do you think?