What do you mean "You don't do dice rolls" ?

[altipueri]

I've seen in a couple of threads about hitting moving targets and other accuracy things the statement along the lines of "we don't do dice rolls, it's all physics."

Er, surely it's all dice rolls and probabilities isn't it?

[Thomm]

I guess the message is that the "dice rolls" are restricted to introducing small random errors in the aiming process (to account for gunners who are not ballistic computers), small random errors representing the dispersion of the weapon itself, and, perhaps some random variations at the armor plate level to account for variations in armor quality.

But once the projectile leaves the barrel it is following a physically correct path to the impact point, and the penetration calculations are performed for the correct plate at the correct impact angle and the correct impact velocity. That's the physics part.

It is certainly more physics based than "dice rolling" schemes for evaluating hit and damage chances, but certainly relies on some randomization, also. This would be true even for "perfect" ballistic computers doing the aiming, as long as the dispersion of the weapons is being taken into account.

But even then numerical errors in the calculation of the flight path will introduce some amount of randomness.

In the end, you may obtain reasonable enough results with a dice-rolling algorithm, but it means work for you to do the necessary abstraction. With a physics based algorithm you can directly input measurable data such as weapon dispersion and couple it to reasonable assumptions regarding the accuracy of gunners. If you get new data ("veteran soldiers should be able to aim better according to study xyz") then you can apply it across the board and do not have to mess with abstracted hit chances.

Best regards,

Thomm

[Battlefront.com]

Thomm has said it well. There's a big difference between using carefully constrained probability to influence an outcome and using probability to represent the process itself.

Firing at a moving target in CMx2 involves a ton of specialized equations which are largely based on real world physics. There is constrained variability within these equations, to the degree necessary, but by and large it's scientific from start to finish. For example, we do take into consideration wind speed and direction, but do we measure the coefficient of friction down to level that would make a molecular scientist impressed? No So we introduce a reasonable approximation for variability within equations to represent such things. But randomness, in the true sense of a die roll, plays no significant part in the process of CMx2 determining a shot being fired at a moving target.

CMx1 had a much smaller suite of specialized equations which were themselves less sophisticated. To make up for the simplification of reality the impact of randomness needed to be higher than for CMx2. Which meant that the end results were not determined as much by physics as luck.

Most wargames, however, are massively dependent on randomness. So even CMx1's approach was far more science based than luck based. If you look at a board wargame, with it's Combat Resolution Tables (CRTs) and dice... it's almost all luck and very little science. It had to be that way because a game like that would be unplayable otherwise.

Computer wargames that follow the "dice" model have their own reasons for doing so. Not much point in having sophisticated physics bases equations when tanks have health bars and power ups For other games it's simply not what they value. The definition of good enough for a mass market "war" game is not the same for a serious minded "wargame".

Steve

[altipueri]

I suppose I've been playing serious wargames then for about fifty years. Back then we had a huge file with all the armour thickness, angle and equivalent thickness data we could gather together with armour penetration data at various distances. I guess that's why I still hanker after those CMx1 tables - they were just like the pages we had on each vehicle and gun.

CM is still by far the best game of this type for WW2 and it is the basing as much as you can on real data that I appreciate.

Luck both good and bad has a much greater bearing on life than people want to accept. There's sometime a burden of feeling that if only you did everything right you could get all your guys across that stream with no casualties.

[Michael Emrys]

Luck both good and bad has a much greater bearing on life than people want to accept.

I very much agree. Still, skillful play allows you to stack the odds in your favor even if it still comes down to a die roll in the end. I'd much rather have a 90% chance than a 10% chance.

Michael