Yeah the problem is we can’t see if it really work cause devs also modified the ai before we can try to set it by our own so i didin’t touch it cause ai is better then before patch
Right now i got little less theb 50% def win and this is enough for me i don’t feel the need to play with the settings
Good point! It would take hours and hours of experimentation to get any feel for how the customization actually affects things, and who has time for that when you are earning nothing in practice battles?
I tryied it once and i think you still win xp and gold
edit: nevermind i just did it and won nothing
Well its a new week and the AI played totally different to last week for me at least anyway.
This week I baited with skulls all week except for when they had a spell to fire.
That trick didn’t work once last week 
AI keeps missing 4-5matches in favour of skulls again didn’t do that last week.
Mana cascade surge’s have been Nuts since update. I agree with Talia parks that console ai is a Mario kart system now and that’s not a good thing.
@slumzee I noticed this too. I also find this incredibly annoying. It seems the AI get some phenomenally lucky matches sometimes… a lot of times. There have been many times that I got pretty upset with the AI’s luck and I remember pre-update I never got this upset with lucky AI. Don’t get me wrong, it still happened, but not to this level though.
Here’s an interesting thought, which is: I wonder how much people think the current console AI is working correctly? Or think it’s currently out of control?
It is far from out of control, but it has never worked correctly. It is nice of them to give us new AI customization options, but when the AI is not properly analyzing board state it is not really helpful.
II’m in agreement with you 110%, yes. The console AI is bugged and needs addressed/looked at. I’ve done a lot of defence testing this week and the results have been severely negative.
Have you noticed how often troops are holding spells instead of casting them? That annoys me to no end. Why is my Dragon Soul and Khorvash holding their spells…? Gem exploders had a crazy high cast priority before the guild update and that’s why nobody who paid attention to AI troop cast priority used the Desert Banner on their defence team with a Gorgotha in it!
@Alpheon could you ask the console team about this please?
Talia had lots of beliefs on the nature of the AI. Based on my experience the is no Mario cart effect. In addition to Mario Talia also wrote half a dozen threads with a dozen posts/videos that the AI would “give up” (that not Mario).
As I have written in the past if you watch closely you will see the AI on occasion will take a non optimal choice, sometimes leading to a better outcome, sometimes not. A person tries to interpret both results and concludes the AI “gives up” or goes “full Mario cart”. Nether is occurring.
Watch closely what the AI recommends for your turn (or what it does on its turn). It will occasionally ignore a 5 match recommending something else. 4 matches… On and on and on, I’m not re typing it again.
The “advanced” console AI still can NOT properly identity all Gem matches. This leads to people creating crazy theories like Mr Strange is out to get them, etc.
Again I prefer console AI to PC any day.
Hey All, we have been keeping a close eye on this thread and your opinions of what is being said. The new AI system is not meant to be perfect, first and foremost. Just to give a small overview of the new AI system, there is a calculation given to each and every board move and spell cast against every combination of spell targets. These final scores are then (slightly) randomized within a threshold, meaning that the best move may or may not be the move targeted.
Why do we not use the best move (meaning occasional ignoring of what you may expect the AI’s move should be)? The main answer has been given above. The AI picking the best move ALL the time, suddenly it is “cheating”, or games take longer. It also makes the opponent act more like a person (not the “best”, but not the “worst” either), where it does make moves different from what you would expect.
With the added addition of the gem preferences for PvP, it was added so you could make your defense team more personal, making the AI slightly closer to the person you are up against. There are some others on the forums who want real time PvP, and slightly customizing the gem preferences for your defense team would make it feel slightly more like person you are up against, making battling the same team different for each person you are up against. Again, this could be seen as a way to make the AI “cheat” or battles take longer, but isn’t it really just the opponent knowing that you may be “baiting for skulls” or using the usual tricks to beat your opponent.
Also, just to point out another thing, there has been no modifications to the gem board or any luck factors in the last few updates. Any “luck” based complaints would have been present in previous versions.
We are looking to improve the AI over time, so threads like these are useful to know how everyone receives this new change.
Well I’m a human and I don’t always make the best move. I often set up the AI or miss a 4 match occasionally, so I totally get what Sirrian is saying. I think it’s pretty impressive and much better than previous.
But you guys changed something? cause i use gorghota often and before patch when i was fighting myself he was always casting his spell and now on the new patch and default setting he react differently
They changed how the troops react, but nothing to do with how the gems drop or how ‘lucky’ the AI gets.
Ha ha I think we have all done a lot of testing this week mate. Mate if they want to run desert banner with a gorgotha team on defense then by all means please do. 
I played around with the sliders for a good 90mins this week b4 I was happy with my team. My defense wins speak for themselves.
What % im curious 
Lol, at work will update tomorrow ricky
Lol no prob right now i am at 78 win 92 lost it is not so bad
My current defence team is at 85 wins and 53 losses, but that’s a mixture of wins from 3 trophy players and 2 trophy players, too.
My defence rate swings so much from day to day (with me changing nothing), that I would never have confidence that anything that I did change had any significant effect. I haven’t changed my team in a couple weeks and I can go from 60% win rate to 0% (with between 10 and 20 matches) on any given day.
Good on you for actually experimenting with this. If you were in practice mode, did you feel like you could actually see the difference as you made changes?
Yeah me too it is random every day i can win 7-8 then lost 10 but for the number of fight i do i consider it for a good win %