How do you improve PzB???

[Bayes]

I agree with some of your analysis ComradeP.

However, here is a counter example to your claim that tank battles are non-brutal in Panzer Battles. When I first played Nepkhaevo - The Meeting Engagement as the Russians, I split my tanks in two groups and got caught at close range in the open. At the end of the scenario, I had lost all of my tanks, and the Germans gained a major victory. Felt pretty brutal :-).

If I simulate a battle between 20 Guard T-34 and 20 PzKw IVg at Range 1 (opportunity fire not simulated), this can also be studied more systematically. Assume that one part breaks off the battle when 5 or less tanks are left. Then 10 000 battles ends in the following amount of rounds:

14% ends in 2 rounds
67% ends in 3 rounds
16% ends in 4 rounds
2% ends in 5 rounds
1% ends in 6 rounds

E.g., in one play the battle could be over in 2 rounds, perhaps feeling brutal, while in another play, the battle could take 6 rounds, perhaps feeling lengthy. I guess the question is whether the overall distribution of outcomes is a rough approximation of what can happen in real-life, and determining this for all aspects of the game probably takes a huge amount of play testing.

Note that in the above scenario, firing first, the PzKw IVgs wins 99.75% of the battles, with 15 remaining tanks on average.

If the battles takes place at Range 6, on the other hand, the low probability of causing kills at long range will make the battles much more lengthy. Then the distribution becomes as follows:

1% ends in 6 rounds
7% ends in 7 rounds
15% ends in 8 rounds
17% ends in 9 rounds
16% ends in 10 rounds
12% ends in 11 rounds
10% ends in 12 rounds
7% ends in 13 rounds
5% ends in 14 rounds
3% ends in 15 rounds
2% ends in 16 rounds
and so on…

with the PzKw IVgs winning 93% of the battles with an average of 13 remaining tanks.

If one would, as Xaver suggests, provide unit types with individual Range Effect Modifiers, one could for instance give the PzKw IVg a modifier of 1.25 instead of 1.5 to reflect superiority at long range. Then the battles at Range 6 end more quickly and the PzKw IVgs win 99.9% of the time:

13% ends in 4 rounds
44% ends in 5 rounds
31% ends in 6 rounds
10% ends in 7 rounds
2% ends in 8 rounds

Here 17 PzKw IVs remain on average, after winning the battle.

As a final comment, addressing ComradePs example of inability to take out T-34 at long range. I do not think rare occurrences should be impossible, but instead have a low probability of happening. As an example, I set up a battle between 5 T-34 and a Tiger at long range in Combat Mission. The game reported that it was more or less impossible for the T-34 to penetrate the armor of the Tiger. Yet, the T-34s won the battle because they managed to damage the gun of the Tiger.

Anyway, I look forward to seeing what David et al comes up with regarding ranged combat. :-)

Bayes