in some was yes. but there are ways, that doesnt effect the game experience that badly.
but this ai reminds me of the ridiculous ai from empire earth (release 2001!) - if you remember, this ai could spam troops over troops, gather resources likes there is no tomorrow and it wasnt fun to play against this ai at all and its remembered as one of the horrible ai´s of any rts.
so there should be any good reason, why any developer should go over 20 years later for such a bad decision, we all know it better.
Then they took those cheats bit by bit.Now it’s a shell of former self that anybody can beat.They didn’t add new logic to make up for the removed cheats.
Everyone else has said it, but I created an account just to say it. Why would you not ADD a fifth difficulty? Replacing the existing one makes no sense. Also, why not make the level the AI cheats at configurable? Seems so obvious. I’m all for intuitive menus and reducing clutter and the developer knows best and all that, but only up to a point. Breaking what worked before because some people want it is not great. The obvious reason: some people don’t want it.
This is a profoundly terrible change. Not only does 1v1 an AI feel insurmountable, this effectively breaks the final challenge of all civ masteries, which is win a 1v1 match against the Hardest AI.
This change is tantamount to a feature being removed. The Hard AI is still too easy, and the Hardest AI now uses megacheats.
There are two possible fixes that I can see:
Add a new AI option above Hardest (this also fixes mastery challenges)
Add a new Handicap option that can be applied to any human or AI player in a custom/skirmish game, altering their resource collection rate at 0.25x intervals. Perhaps it’s best if a good human player has a 0.75x resource collection rate. Perhaps its best if a Hard AI has a 1.25x resource collection rate. Players can figure it out based on their preference.
I have been having a lot of fun in my household introducing my partner to the game. She is a casual player and has worked up to beating the Hardest AI. I am afraid she will quit the game over this change, she is seeing me struggle against the Hardest AI and I have played RTS games at a high level for over 10 years.
The masteries are easy. Set to starting age IV, max resources. One villager makes houses unless you’re Mongols. The others make siege workshops, then make bombard cannons as fast as you can. 15 should be plenty, double click one of them and right click on the AI’s starting TC. GG in under 4 minutes (mountain clearing is a good map to use, less distance to get to the AI’s TC).
Adding my 2c… would really prefer cheating AI to be a separate AI level, so leave “Hardest” alone and add “Cheats” as new difficulty rather than replacing. Nice to add new difficulty for those seeking that challenge, but very disappointed to see a challenge level disappear. Please change. Thanks.
Hope they bring back the old AI for the Hardest difficulty for casual players like me. agree with the solutions presented by others, just add another AI difficulty “Epic” or “Insane” or make handicap a checkable option for those who wants an extra challenge. please dev make it happen
Just make men-at-arms, the AI doesn’t know its counter is the knight. or in general, he does not know that he should produce units depending on what the rival produces
I tried making men at arms and it seemed to have no problem working out that if it simply makes more men at arms than me, its greater number of men at arms counters my men at arms AND it can still afford to age up while I’m having to go all-in castle age to have any hope of matching its production.
Having watched the first game he plays there, obviously we already know he’s a far better player than me, he beats it because he’s better at the game than me, fair enough.
But it also highlights that what we have here is a cheating AI that is still absolutely braindead. It would be much better if it at least fixed the most obviously braindead behaviours. It’s absurd that he is able to shut down its eco with just a single scout at a collection point combined with his high apm to micro them as well as manage his base. And then having seen him make 100% men at arms it still sends loads of spears. And it repeatedly suicides its army fighting under his TC instead of doing the one thing it should excel at, which would be to send some units to each resource collection point and use the insane apm it ought to have to harass the heck out of his villagers.
Edit: And the second game even worse, he runs a scout in circles around a tower and the braindead AI chases it with all its spears till all its spears die in the arrow fire. These are the things they need to fix before looking at making it cheat with extra resources.
I personally have no interest in playing the AI in these ways because there’s no long term interest from doing it. I’ve done it before, e.g. here is a way to beat the AoE II DE extreme AI because in this specific scenario it is also utterly braindead.
It’s kind of fun to work out these exploits, another one I did was trushing it on hideout and building houses in front of each gate. It’s too braindead to understand it can just delete a section of wall to get out. But there’s no long term interest for the player base from these things, there aren’t enough of these exploits, so you work a few out, have fun winning a few games, and move on to another game.
Beasty’s +/- the best player in the game, but I feel the AI just failed in the first game. Sure its exactly the comment he called out that end - but I still think it’s true. He says he kept them off gold - but he didn’t really. The AI just didn’t put almost any workers on gold. It made traders I guess, so maybe that was its “internal logic”, but it didn’t have to do that.
But then we see from Beasty’s AI tournament (i.e. today’s stream if anyone watched) that sometimes the AI seems to just go “not feeling it today”. So I kind of feel that’s what happened here.
When I play it I get something much more like Drongo. The AI sticks maybe 5~ workers on gold from 4~ minutes into the game, and now has the resources to make Knights from 2 stables continuously (7 workers will let it buy upgrades and eventually go to Castle). Net result it can have made about 17~ Knights by the 11 minute mark (i.e. Drongo’s 2nd game) - not five.
But this seems to be a bit of a thing. I’ve seen people say “I’ve defeated the AI with 3 harassing longbow’s straight from Council hall”. Well good for you - but in my games by the time your 3 longbows get there, the AI will have half a dozen units and just jump them. And then it makes a lot more.
it seems that there was no test of this AI, it seems pure laziness, a simple change to change the collection levels would solve it. This AI demonstrates the disregard for game users/consumers.
It’s probably a timing difference. It might only take e.g. 15 seconds to make all the difference. In the 3 games I’ve seen so far, 2 from Beasty and 1 from Bee, all 3 have had some form of unusually early attack. I can’t try to copy Bee’s as you can’t do it in the normal game, and the pup crashes for me. I can’t copy Beasty’s as I’m not good enough to do the required micro at the enemy base at the same time as optimally managing my own base. As I said before, it’s fair enough that he’s winning because he’s doing things he can only do because he’s a much better player, but it’s not going to provide much interest for most players if the only way to beat the hardest AI is with a small number of options to make the earliest possible attack. And if they ever make the AI better at dealing with those very early attacks, even the pros might not be able to beat it any more.
