I know the "Heroes" series worked extensively in this area, so if your really interested try out heroes 5 for example to see how they did, had a fantastic AI compared to it being an AI.  It should really put these decisions into a multiple formula answering thingie, if I have more archer power than the enemy, try making my melee's form a chokepoint. - If I have less archerpower than the enemy, move melee's forward in a sensible manner, etc.

Sincerely
~ Kongdej

I think archers/ranged units should be what alleviates this scenario. 

I love that we're now at a point with this game's development where the most often mentioned critiques within a thread are those concerning sound effects.  Excellent work!  This has been an immensely fun gaming experience so far (notwithstanding the first few days after WoM came out, obviously).

yeah when the major issues are sound effects and fonts, you know you're on the right track.

I don't follow your logic at all. A stalemate because noone wants to suffer first strike would only occur if you are fighting 1 unit vs 1 unit. As soon as you have more than 1 unit on each side, the stalemate would never occur if the AI played it smart.

What the AI currently does:

In the above screenshot, it moved all 3 red units into attack range. The human player will land 3 first strikes and the AI will lose.

What the AI should be doing:

The human player now is forced to focus all its first strikes on a single red unit (preferably a tank unit or an expendable unit such as militia). The following moves the AI can use its two undamaged units to attack because the human units are now in attack range AND THEY ARE NOT AUTO-DEFENDING making them squishy.

This is just an extremely crude example of why doing what Frogboy does is bad, and makes for a shitty tactical AI. I've been saying this since early E:wom days and it remains unchanged.

The AI should be at the level of chess engines when it comes to deciding moves. And by that I mean it should be able to think 10 moves ahead with ease. Instead it has a very simplistic "target lowest hitpoint unit, attack/move/spell" system that makes tactical battles boring and easy to game.

Seriously, even if I'm fighting 5-6 units vs 10 AI units, it is rare that I suffer a single unit lost. In 80% of the scenarios NONE OF MY UNITS ARE EVEN DAMAGED at the end of battle. So that's what is going on with tactical AI. It's bad.

Depends on the sound effects
That said, I feel the game have done good, now its not perfect (to me) but that is why they continue to develop it

Sincerely
~ Kongdej

Is there going to be a part 2?

Well said Hf. I suspect beta 5 will contain major updates to tactical AI. From a developer standpoint, you would want all the bells and whistles in before you start making the music.

AI will be the priority of beta 5. I really hope...

"Heavenfall Reply #50: (Imagine all the text here)"

Thats a fantastic way to explain one of the the tactics to use!
Thanks for explaining that, Personally I hope the AI also just use a slightly "less perfect" solution, just to surprise the player.

Sincerely
~ Kongdej

Except he has done similar stuff in every video he has posted, the AI did the same thing in the video, and the AI does the same thing in the current version. Plenty of evidence to support the theory that he did not just fool around to make the video shorter.

HF, the human player is NOT forced to attack the unit.  You can DEFEND once again (or move backwards) if it's not in your best interest to attack.  So if the AI is waiting for you to make the first move and you never do, you get a STALL!  I'm all for the AI making tactically intelligent moves, but I don't want the AI creating stall scenarios.  The AI must come at the player ... as tactically intelligent as possible, but it must come!

The current problem I see with having 'intelligent' combat is that it currently doesn't exist. As is, you gain the most benefit for striking first. Hense, even if the computer tries to position for first strike, you just move back further.

It wouldn't be fair to the player if your tatics are penalized with a 'stall' and stalling should be considered legal moves.

Rather, I believe the solution is in game balancing. First off, and this may or may not be true as I have not tinkered with them yet....but pikemen or halbrieders should have some major benefit to being part of your army. Just like some weapons get backswing or maul, reach weapons should have, or be given, an ability that makes the player WANT to include them to take the first hits from your opponent.

Maybe have a 'first strike' ability attached to both the pike and the halbread, but have the pikes give additional bonus against mounted units and halbreids give additional bonus against foot-units. This way, you can position some of your troops to take the first hit, but still gain first strike...and then have your other units to obtain the tatical edge in movement or strength.

Maybe this is too complex and there should only be one type of weapon with this bonus type...that's for the developers to decide. However.....there is also the 'lance' to consider...mounted with reach can go a long ways. (sorry....bad play on words).

I thought spearmen were immune to counter attacks (at least with swords and such)?

Incorrect. The benefit of striking first is only the best option when you specialize in that tactic.

You can build a unit that is meant to be hit first for an advantage. The Kraxis mounted spearman will do better in a line, waiting for the enemy to make the first attack. Put some mages behind those units and sit back. Altar too has a great shortsword that is better for defense. Put a the swordsman in the middle of a big group and let him tank. With the right traits and armor, you will get 2 counterattacks and take almost no damage. Your damage per turn is still very good though. If you don't see the benefits of letting the enemy come first, you might want to play some more without ever attacking first. It doesn't work for every faction, but it will for Gilden, Altar, Kraxis, and Yithril.