Im Westen nichts Neues (All Quiet on the Western Front). Watched the german Netflix release (with subtitles). Had seen the 1930 and 1979 version. Good remake and highly recommend this film.

Dune (Frank Herbert)

From what I read this happened after Waterloo and Austerlitz with bones from both man and horses... (Bone meal)

Another aspect of AI in strategy games would be options to have different styles of opponents (instead of difficulty levels). More aggressive or defensive play styles for example.


I'm all for this. And would you believe it, there are games with such feature. Some of Sid Meier's Civilization ...

Likewise, wargamers can talk about AI.


Likewise, non-programmers can talk about machine-learning for computer opponents. Uh-huh. :roll:


Likewise, non-mechanics can talk about car engines ;-)

I'm not a machine-learning or game developer (though I am a professional developer). I was just ...

sPzAbt653 wrote: Sat Aug 06, 2022 1:52 pm Playing an AI that doesn't know the rules sounds quite messy.

To be clear the AI wouldn’t train/learn on your machine. By the time the game is released the resulting AI knows how to play the game.

Yes, that is what I was thinking of. Some future version of machine-learning. Of course, the level of complexity for strategy games is far greater than for Chess or Go. So, such a system might not be feasible in my lifetime (and too costly for the game industry).

To make such an opponent ‘enjoyable ...

I wonder if in the future wargaming AI could become trained instead of programmed. And not make use of rules or information that are any different than those available to the human player.

Programs like MuZero train without knowing the rules.

Not only could such an AI be more challenging than most ...

One extra would be to make the image of the map exportable. Would help a lot for AA reports.

I doubt the game can be modded that way. For example the air war is based around planning and execution for a week. Same goes for all the logistics math I assume under the hood...

I would love a tool like that, to calculate how long it takes (without battles) to get from point A to B (by marching or driving). This would help with long term planning if you play with movement fog of war on. And would help to find the fastest theoretical road from point A to B.

Right now I use a word document to keep track of the narative/history of my game. I first make a one page operation plan, that looks something like this:

O P E R A T I O N B A R B A R O S S A

Start date: 22 June 1941
Planned end date: winter 1941
Objectives: Destroy Soviet/Red Army and form a ...

Other map modes could for example be rail or road mode. In those modes show all the hexes in the same color (except blue for water). That way you would have a clearer view of the transport network.

Great job, thx

+1

Yes terrain would be solid colors.

Ground state would be solid colors too for clear, light mud, heavy mud... (proper term would probably be ground condition).

<t> I would suggest following, have map modes. For example:<br/>
- current map mode a s standard<br/>
- simplefied terrain view<br/>
- simplefied ground state view<br/>
...<br/>
<br/>
A bit like all the map modes you have in for example EU4. This could help with planning and make it easier to ...

<t> Been thinking about this too, mostly for major rivers and road movement. For rail bridges I kind of assume that they are part of the rail damage/repair system.<br/>
<br/>
Maybe a number of bridges that are in a hex that can carry tanks over the major river in a given direction... An option to ...

Once again great work cblattmann, thx

<r> I used the template for a luftwaffe only banner (I think there are only about 10 of them in the game).<br/>
<br/>
<IMG src="https://www.matrixgames.com/forums/upfiles/72589/3EFB13DA01844F8FA61A17C94CF615D0.jpg"><s>[img]</s><URL url="https://www.matrixgames.com/forums/upfiles/72589 ...