One little detail about NCP2.

[Johan]

NCP2 is the first game me and Philippe Thibault are working together on for over a decade.

What was the previous game we made together?

Europa Universalis I...

Just saying... Revolutionary things in the game industry has happened before when we have worked together...

 

[Johan]

In the turn/real time debate.

Turnbased gives greater abstractions, and you can't make a realistic depiction of warfare.

Without real time, you lose granularity unless you have very very low turn sizes.

 

[King]

Is AI performance better in turn-based games? I don't have real knowledge about this, but it seems intuitively to me that it could make things more manageable for the AI (more time for process the 'answers', and, a priori, easier to cope all relevant variables with correspondant algorithms).

I feel walking on thin ice in this...

Surprisinly enough the answer is no. The biggest problem with the evaluation logic of a turnbased AI is not just what I am going to do but to be really good you also need to evaluate possible options that other players will do in their turns and how you are going to counter it. A Chess AI can be very good at what it does but it really does take a while when you crank up the level. In a real time AI you can react there and then to what the opponent does. The challenge with a real time AI is when to be stubborn and when to react.

 

[Johan]

Then why, for me, does warfare in most TB strategy games feels more realistic than in most RT games? Is abstraction something bad?

Its because of what you think a wargame should be about.

Combat Mission is a good example of a real time strategy game with "forced breaks". If it had been a turned-based game, it would have been far more abstracted.