I wasn't attempting to beat my chest in declaration that I am some super, AI defeating machine; When I said that there was no challenge, I didn't mean that was due to some legendary strategic and tactical skills that I posses. I literally meant that the strategic and tactical AI poses no threat: the strategic AI commands an empire that has a power rating 5x higher than that of my own - they declare war on me - 100 turns later, I have yet to even see a single attack made by them. The tactical AI seems to suffer from the strongest case of tunnel vision I've ever seen - it will focus all units on attacking a single enemy unit until the battle has ended regardless of how the enemy changes position or attacks.

I'm looking forward to the 1.2 beta and will post my experience with the AI after a game or two.

But that is gaming the AI. I do see a point being made that you are only following a logical course of action by retreating a wounded unit, but at some point you just have to do the honorable thing and let he or she die in the name of the Sov. Otherwise the game is too simple to beat. I am just saying that computer sparring requires humans to give and take when we find an exploitable glitch in its logic. 

Not that you are in any way being coerced by my words, just my way of seeing the world. 

I believe that the upcoming release v1.9 will fix the invincible armor issues as well as remove the "Blink" tactic, making Ridiculous a potent setting.  Sure, someone will find a trick to defeat it easier than most but I am looking forward to this release as my original release was 1.1 as in I bought in December.  I have played about 11 games all the way through (Challenging world, Extreme or Ridiculous AI) I had to start over twice in 2 of those games as an 8,0 Outlaw defeated my Zanusk on season 6 (terrible damage rolls).  The other time I misclicked on my hero during a fight and lost to 2 Outlaws on season 13 (i.e. before armor).

I am patient and can wait for the release.  Thanks to Brad and company for their continued support, feedback and overall information.

V1.19

The AI being unable to cast most spells in combat... nah, that's not decent.

(It's a bug ofc, I know. )

I have to agree with that the AI isn't too bad. But there are plenty of "loopholes" for which the AI has no answers. Mainly rushing military tech and power inflation of units. You've once said, that there is no way to code the AI so that it can answer to rushs. That is true I'd say. Which means the game needs to be changed in a way, to make tactics which are way too powerful either very hard to execute or even impossible.

As it is, the AI can't deal with the power inflation given by military research (it doesn't equip its champions with new stuff, doesn't phase out obsoleted units with better ones, doesn't factor it in deciding for war) for example.

Real exploits, like all terraforming spells, blink, diplomatic capital or trading for example don't need to be used, when you can defeat an AI since it can't deal with the power inflation given by research. And while this imo is more of a design flaw then an AI flaw in itself, it means, like 1upt in CivV, when you have something fundamental in the game which the AI can't cope it, your AI will be seen as weak. It may not even be possible (possible as in time and money being restraints) to code an AI that handles those well, but then game design has to change to accomodate that.

/demandobot 3456B

Well said  Frogboy, there is really nothing as sad as the SW fanboys raging at Phantom Menance at the cons.  Your childhood was pimped out for dirty nasty things, move on to better material.  I think the best part of this forum is that it consists mainly of adults wanting to have a conversation about a shared hobby.

WHAT I SPECIFICALLY WANT ==

It would be fantastic if the modders could get a short write up of how to use the new tactical AI tags that Charles has added.  For special abilities and spells we have created, knowing how to make them play well with the new system would save us a lot of time on trial and error.  This is particularly true for mods that are meant to extend the vanilla game via spells, as getting the AI to use them tactically has previously been extremely difficult. 

(I have the screenshots to prove the AI uses at least some of the Strategic spells well already)

Thanks for posting and continuing to deal  with the 5% of people who seem driven to suck.

Frogboy, your killing me. 

and, more importantly, anti-ACNE cream.

Took me awhile to stop laughing at that. 

I happen to gree with you and HF about the improvements made to the AI. Does it still have issues? Yes. But they are much improved on. And more importantly, all of your work is moving in the right direction.

I also must admit that by using several of the mods available, it has improved the game even more, and i quite enjoy playing it.

Keep up the good work. You have my support.

I got my butt kicked by the AI on novice today...oh the shame   , was my own fault for being reckless with my soverign early on but still the strategic AI is plenty good enough for me.

I specifically want TAC battles in Multiplayer.

I specifically want an iniative system so my archers don't shoot the same unit six times in a row every time it takes a turn.

I also specifically want a method to repurpose my peasants with sticks to be peasants with bows without having to serve them up as hors d'oeuvres to a nearby dragon.

I also specifically want to be able to take 3 peasants of similiar design and make them a group, or take 5 peasants of similar design and make them an army.  Or take an army and split them up as 5 independent peasants of similiar design.

I believe I specifically mentioned these several times.

I want this to take a much higher priority than sparkles, glitter, and other superficial, pretty "fluff" that has been the UI changes and graphics enhancements.  I also want this to happen as part of a patch so I can play the broken, ugly game I have now instead of having to wait for a whole new game that was loosely based upon what this one should have been.

Don't get me wrong, the new pretty stuff is nice, but now all I have is a pretty broken game instead of an ugly broken game.  The pretty stuff was definitely needed, eventually.  The core of the game should have been fixed first, the real problems addressed first, the hard questions answered first, before you slapped some make-up on a horse and sent her off to the pageant.

Call me a rude, hyper-entitled demandobot if you want, these strike me as functional changes that should be in the existing game that was far more important than the UI-porn the last few patches gave us, and definitely shouldn't count as some of the "new content" that Fallen Empress seems to be promising us. 

The 1.19 patch fix didn't change the current situation in any noticeable manner, so the ridiculous AI still isn't that hard to beat if you know what you're doing. Until we get a much stronger rebalancing patch, it'll stay the same. Unless the AI starts using cheesy tactics, which would not be the right fix for this :/.

And I don't want this.

There now your request is canceled out by mine.

@ Frogboy: Still would be nice if easily reproducable gamebreaking (and by that I mean easily fixable gamekillers) exploits (as in the AI can't use them and has no way to defend against them / block them and they work in nearly any game) which seem easy to change / axe get adressed as we report them instead of being seemingly ignored (posting in Elemental Support and the changelog-threads about the issue since the 1.09 Beta, no response yet...):



To give the 2 most glaring examples:

Diplomatic Capital:
in the treaties-part of the negotiations-table being worth 5 times more then elsewhere (like in general trade) meaning a mediocre amount of DC is enough to rob any AI blind (aka steal all its resources) by going to treaties and cashing in the DC and then going to general trade and trading all the gold back for more DC and repeat the cycle until the AI is piss-poor.
There are a number of ways to fix that (including solutions requiring a bit more thought to keep the nice concept of DC being worth more for treaties ect.).

Trading Heroes:
A single hero might be enough to rob the AI blind since trading a Hero to the AI is worth more then trading him back right after. Doesn't even take long to rob the AI blind this way. The numbers on the table even show the difference in value so its obvious something is off (either the players heroes are overvalued or the AIs heroes are undervalued). Value seems to be affected by recruit-price (so bards, Cha ect...) and relations as well as AIs recruit price if I'm not utterly mistaken (so to reproduce you might need a high-charisma and bardic souvereign.).

If it got changed in 1.19 fine but sadly that's not documented. (anyother gripe given there is a number of undocumented changes like coming of age in the dynasty which lends credence to the impression that internally not every change the programmers do are even communicated to the internal QA.)

For all other goods this kind of blind-robbing is prevented by AI goods being worth more than player-goods on all counts. Which sounds sensible (at least I haven't found anything simmilar otherwhise I would have reported it already.



Mainly by this post I mean the missing feedback given the gravity of the exploits.
A simple: "It's on the radar" might help.
That you indeed take note of such reports is visible in the bold axing of blink. Which sounded rather necessary if rather sad..
Repeating these observations and how to replicate them in nearly any post made about the topic didn't seem to help to get it on the dev-teams radar (And to be honest my time and energy have been severely limited until very recently do to private and professional time constraints. Thanks to the fast changes and patch barrage (Kudos for that btw. ) I keep posting and testing but Its sometimes hard to motivate myself doing that to be honest... ).






Also to be honest putting the blame of getting the AI to the appropriate level (which I for one do since I noted the difference of normal AI in my first 1.1 game without reading about in the forums since the difference is indeed glaringly obvious to anyone regularly playing on riddiculous) on the players is not least the fault of whoever changed the default AI to normal instead of world-difficulty.


If you want players to provide constructive and non-confusing feedback, better ought to make it easier for players not to fall victim to those kind of pitfalls. (and this pitfall is a well hidden highway wide-abyss you have to actively avoid as a player...)
It has been different before and a sensible reason of a change of this default is beyond my understanding. (Can't say I find a response in the way of: "Fools why don't you mind the AI difficulty is different than world dificulty." (a bit overstated given ) Very helpful for your goal of getting useful feedback.)
I work in ISP inbound tech support at the moment and falling victim to "ivory tower elitism" of expecting your customer base to have your degree of knowledge just ought to complicate things tenfold...
Not every player / customer is an Programmer / UI-designer and a minor note in a changelog isn't enough to catch everyone's attention.

The "stupid customer approach" is bound to poison any kind of debate like that on top of putting workload on your plate without any discernable advantage to you (unless of course you like to complain loudly how inept the players/customers are for missing something "so obvious" A common theme at where I work but that doesn't help fix any problem except perhaps creating a false sense of self esteem and superiority. I doubt that is your cup of tea to be honest. ).
(I could strangle those misguided designers who write hard to read /miswritten Wireless-keys under certain Routers... )

Simple enough fix: Change the default of AI difficulty back to world difficulty (making differences of AI and world-difficulty opt-in instead of opt-out) and you will get lots of more useful feedback with less time for you to find out if the player did it wrong. Promise.

I love you Frogboy

If I was of different gender and not married I would propose this moment

On a serious note, I agree with you. I bought this game, and it looks like I will not play (but will play the Fallen Enchantress as it looks like all the things you did wrong with Elemental you will try to fix there), because of posts like this from you. This gives me faith that one day my investment will pay off

Stardock folks, thank you this update

Here is some feedback from someone who played from Aug-Sept and then put the game on the shelf.

Played my first game since then on Sunday, was very suprised how much for the positive has changed.  Even on regular setting AI gave me a couple of suprises, the game actually has some challenge in it now.  Will try with the harder AI settings.

Thank you for the work you have put into the game the last 4-5 months, it shows.