about pre-rolled die rolls

[cranston]

I have a question regarding the game using pre-rolled die rolls.
More specifically, how many die rolls are saved? And when were they generated?

I was testing this by playing a skirmish game against the computer.
I saved the game just before contact, played a couple of full turns (I-go, AI-go, I-go, AI-go), took screenshots of results of each clash, reloaded, and did all my actions exactly the same.
Did this 3 times, and replicated the same results 3 times.
Casualty numbers exactly the same. Units disrupting and fragmenting exactly the same. Fall backs exactly the same.

I then changed thing up a bit by changing the sequence of the units I moved in turn 2.
I.e. Attacked with unit B first instead of with unit A.
The results then changed, but repeating the new sequence of moves consistently gives the exact same new sequence of results.
So I figured it was just the same die roll result being applied with different modifiers.

[rbodleyscott]

Random number generators throw out a series of numbers that simulate a random sequence, which behaves like true random numbers for all practical purposes. However, the sequence is in fact not random at all, the generated numbers will always follow the same sequence if they start with the same seed number. The random number generator seed number is saved when the game is saved, so every time you load the saved game, the random number generator starts its sequence at the same point. If you do the exact same sequence of things, the results will be the same, if you change the sequence of actions the results will be different, as different random numbers from the sequence are used for different purposes.

The only rolls that are pre-rolled are Morale rolls, which are rolled at the start of a side's turn for their own turn and the following enemy turn for each of shooting, impact close combat and melee close combat. The reason they are saved is so that multiple tests can be taken as they arise, but because the random roll remains the same, the overall result will be the same as if all the shooting tests (for example) were saved up until all the shooting has been done for the turn, and then taken only once, as happens in the tabletop game. If each morale test was rolled for separately, the failure rate would be much higher than intended. Apart from shooting and melee, morale tests taken for other purposes (such as seeing a friendly unit routing, or being charged when already fragmented) use a fresh random roll for each test.

[cranston]

The only rolls that are pre-rolled are Morale rolls, which are rolled at the start of a side's turn for their own turn and the following enemy turn for each of shooting, impact close combat and melee close combat.

I don't understand this part.
This sounds as if just one morale roll (before modifiers) is being applied to the entire turn.
Is this correct?

[rbodleyscott]

The only rolls that are pre-rolled are Morale rolls, which are rolled at the start of a side's turn for their own turn and the following enemy turn for each of shooting, impact close combat and melee close combat.

I don't understand this part.
This sounds as if just one morale roll (before modifiers) is being applied to the entire turn.
Is this correct?

No, there are separate morale rolls each turn for each of shooting, impact close combat and melee close combat. Also completely separate morale throws for every morale test due to other causes, but these are not pre-rolled.

The significance of all this is that (for example) if a unit passes a shooting morale test, it can only fail another test from further shooting this turn if it has accumulated more modifiers. Also, regardless of random roll, if it has failed a shooting morale test this turn, it cannot fail another shooting morale test this turn because only one cohesion drop is allowed per turn as a result of shooting.

Similar rules apply to close combat except that a unit can drop two cohesion levels if it loses (overall) badly enough.

[cranston]

I think I understand.
This is one roll for all units affected during a phase, rather than each unit getting an individual roll.
This would explain some things.
On one recent occasion, I had 2 separate units of cavalry take shooting damage, and both became disrupted.
On another recent occasion, I had 2 separate units of yari ashigaru on opposite ends of my line charged by yari samurai, and both became disrupted.
So, the reason for this is due to one bad roll at the start of the phase? And not 2 separate bad rolls?
(I mean, it could've been a coincidence and I got 2 bad rolls consecutively, but it just happens a bit too often.)

[rbodleyscott]

I think I understand.
This is one roll for all units affected during a phase, rather than each unit getting an individual roll.

No. Each unit gets its own separate rolls for morale tests for each of the following causes of testing: shooting, impact close combat and melee close combat.

This would explain some things.
On one recent occasion, I had 2 separate units of cavalry take shooting damage, and both became disrupted.
On another recent occasion, I had 2 separate units of yari ashigaru on opposite ends of my line charged by yari samurai, and both became disrupted.
So, the reason for this is due to one bad roll at the start of the phase? And not 2 separate bad rolls?

No, two separate bad rolls.

(I mean, it could've been a coincidence and I got 2 bad rolls consecutively, but it just happens a bit too often.)

That is the way human perception works. Random events tend to be clumped. If they were evenly spread they wouldn't be random. We tend to notice the clumps of bad luck more than the clumps of good luck.

[cranston]

I think I understand.
This is one roll for all units affected during a phase, rather than each unit getting an individual roll.

No. Each unit gets its own separate rolls for morale tests for each of the following causes of testing: shooting, impact close combat and melee close combat.

Tying the above in with

The only rolls that are pre-rolled are Morale rolls

and

Random number generators throw out a series of numbers that simulate a random sequence

Is it accurate to say that at the start of each phase, the start of a random number sequence (the seed number? the formula?) is generated and saved for morale tests?

Or are morale rolls generated for every single unit, 1 roll each, at the start of each phase.
(So if both sides have a total of 30 units on the field, 30 separate morale rolls are generated, tagged to a specific unit, and saved.)

[rbodleyscott]

I think I understand.
This is one roll for all units affected during a phase, rather than each unit getting an individual roll.

No. Each unit gets its own separate rolls for morale tests for each of the following causes of testing: shooting, impact close combat and melee close combat.

Tying the above in with

The only rolls that are pre-rolled are Morale rolls

and

Random number generators throw out a series of numbers that simulate a random sequence

Is it accurate to say that at the start of each phase, the start of a random number sequence (the seed number? the formula?) is generated and saved for morale tests?

No. The actual roll is saved.

Or are morale rolls generated for every single unit, 1 roll each, at the start of each phase.
(So if both sides have a total of 30 units on the field, 30 separate morale rolls are generated, tagged to a specific unit, and saved.)

No, there are 3 saved morale rolls per turn per unit (shooting, impact close combat, melee close combat), so that would be 90 rolls.

[cranston]

I understand. Thanks for the info.