Faction Differentiation and City Management (more decisions towards specializations) are important to me. But AI Development is tops. I admire the work that's already been done. I'd like to see it taken further. And some of the faction differentiation I'd like to see, involves the AI more fully playing the existing faction flavor. 

Though I do have some thoughts as to how to handle random events; which also provides another delivery method for unique spells, items, and units. As well as adding to diplomacy. It involves reviving the Neutrals as specialized societies. It would likely require a sizable chunk of that N resource. I might favor dev time spent on something like that... which impacts a variety of aspects of the game, revives a lost feature, adds a unique element, and fosters the living world concept of decision based consequence and opportunity. 

I would have voted city management If it had existed.

I voted for the Developers to expend more effort on Performance.  As one of the poor folk who is still stuck with XP (and likely to remain so for the foreseeable future) I have a vested interest in hoping that you will make a big effort to program Fallen Enchantress to run smoothly and quickly (or at least as quickly as possible) on the more primitive platform.  

However, I think the point is pretty valid overall.  Once you sell FE to all the power users, modders, hard-core wargamers, and hard-core strategy (including 4X fan) gamers, you will still want to reach NEW consumers, and more casual PC gamers.  I am sure, ALREADY, that this game is going to be hugely successful with your core audience of Serious Gamers.  Reaching a wider audience should also be one of your major goals, and those folks don't always have the best equipment.

I have actually played the FE Beta 2 for about 20 hours in the past week.  It was a good week for me (free-time-wise); but the shocking reason for my playing so much, was that the basic game looked so totally inviting.  It impressed me greatly!  This is a game with stunning potential! 

That said, I have two other pleas to make.  First, please don't expend so much effort on making a "crushing" AI, that you neglect all the other aspects of the game.  I understand that Mr. Wardell has a particular interest in AI; but I hope his personal interest in this one aspect won't siphon off useful resources, that could be better spent on other aspects of the game.  Second, please make sure that your game has at least one or two fairly "easy" levels.  

Over the years I have played a number of 4X games that had 5 or 7 or more levels, but I usually end up playing on the easier ones.  (I remember one game - from a different company - that had NO genuinely easy levels.  I came to hate that game; and I eventually resolved to stop playing it, and to think twice before ever buying another game from that company.)  

I believe that for every hard-core player, who only wants to be "challenged", to play at the "ridiculous/suicidal/masochist" level, you can have: 3, or 5, or even 10 players, who just want to play at an easy/fun/casual level. 

I realize that you have to appeal to the hard-core gamers; but you don't want to write off the larger number of non-hard-core players.  The solution seems simple enough to me: create 5 - 10 "impossible-to-insane" levels for the former folks, but just don't forget to include 1-3 levels for the latter folks too.  You really don't have to shoot for "lowest common denominator" -- but you really should aim for SEVERAL useful, POPULAR, denominators. 

I love what I see so far!  Please continue to improve performance in so far as practicable ... and aim for MULTIPLE audiences.  Succeeding at those goals would constitute a seriously major achievement!  I also think that would guarantee a highly profitable venture for your company!

Win-win !

City management and then faction differentiation

This.

Ron Paul

If I had to give out scores for what is more important for improvement ...

well ...

Faction Differntiation  (98)

Better AI   (97)

Tactical Battles (50)

Everything else a 20 or below.

Therefore if I had to pick one it would be Faction differentiation, but if I had 2, AI would ALWAYS get my second vote.

(If it were 0.77, I would probably be voting for tactical battles, but those have become LEAGUES better already )

I really would like to see work on making the game re-playable.  Every time I play it feels the same.  There have to be more obstacles for players to overcome that require different approaches.  I don't want any of the tech lines to be mandatory.  If I have to take the same research path at the start of each game, it is just busywork.  We should start with basic techs and have the tech tree allow us to specialize and make our experience unique.

I'd also like some simple improvements to tactical combat.  Whether it be facings, fighting in a line, terrain, obstacles to LOS, objectives, etc.. to make tactical combat a little more interesting.  I do think it is definitely headed in the right direction.

Worth repeating

Tactical battles definitely need to change considerably to be more realistic.

Particularly I vote for devs doing this https://forums.elementalgame.com/418739

Faction differentiation is at the top of my list. But I'm not talking about rules mechanics, they're fairly solid there. I'm talking visuals. Every Empire is the same, every Kingdom is the same. Even if every faction just has an armor set, some weapons, or a unique city hub design, that'd be a step forward. Instead of looking exactly the same.

There was no "tech tree" option, so I voted city management.

I think city management is getting a bigger reaction than it ought to have because the UI places a bigger emphasis on it than Stardock tends to. When you can no longer select the tiles that buildings go on, it will feel like a smaller experience and it won't feel so empty. That's not to say this is necessarily a good thing for players in itself, but I do think it doesn't need to be an overwhelming focus.

I voted tactical battles. But it's not really the battles in themselves but the mechanics behind : accuracy and dodge don't scale, magic damage don't scale with the user (aka Int), too few situational bonus/malus (terrain, in close combat, archery penalty), the -6 init on bows would have a sense if bows were powerfull, tweak init so no unit can play 3 times before anyone, the units don't have any interesting abilities before long, weapons like spear are way too effective (66% armor reduction isn't a bad choice, it's just that there should be an initiative and accuracy penalty) for their cost, the damage output and the hp don't scale, the close combat attack should always get a simultaneous counter attack.

There's no point putting too much work in AI when tactical battles are unbalanced due to those problems.

Line of Sight

Adding Line of Sight rules to tactical battles would increase the importance of terrain and the positioning of ranged combat units (archers and spell casters). By careful postioning you can place enemy units in your line of sight while hiding from view some of your units.  With line of sight you will not know if an enemy unit awaits behind the ridge or treeline until those tiles are in your line of sight.

If you have ranged units may want to engage the enemy when they are in clear terrain where the tactical battle field will likely not block your line of sight. If the enemy has ranged units and you don't you may find it advantageous to engage them in forest tiles where a forested tactical battle field will limit their archers line of sight.

Ranged Attacks - Melee and Magical

From Decision Games Ancient Battles Rules

10.2 Line of Sight (LOS)

10.21 To determine if a missile unit may fire at a particular hex, a straight path of clear hexes between the firing unit’s hex and the target unit’s hex (but not counting the firing unit’s own hex or the target unit’s own hex) must be demonstrable. If any of the hexes intervening between the firing hex and the target hex contains any terrain feature, the target hex may not be attacked by that missile unit.

10.22 If the LOS is congruent to a hexside (that is, falls exactly between two connected hexes) it’s blocked only if both hexes on either side of that hexside include a terrain feature.

10.23 Note that slope hexes are considered clear hexes except when intersected by a ridge symbol between the firing unit and the target.

10.24 The terrain(s) in the target hex and the hex of the firing unit don’t block Line of Sight.

10.25 Missile units may fire through (“over”) other units, enemy or friendly.

Unit Visibility - All Normal Sized Units

If a normal sized enemy unit is not in the line of sight of any of your units you cannot see that unit. Thus players can hide normal sized units behind ridges or behind forests and lure the opponent into an ambush with a visible unit.

Large creatures such as Demons and Dragons are not hidden by terrain.

Forest Terrain Effects

The cost of moving through forest terrain tiles is 2x normal for foot units and 3x normal for mounted units

Forest tiles provide a +5 dodge bonus vs ranged attacks

Forest tiles block line of sight. The terrain in the target tile and the tile of the selected unit does not block Line of Sight.