For a forum that prides itself on trying new things, for the love of all that is novel and interesting, please can we have a new QotM? - Krill

Create an account  

 
Civ 6 Release and Update Discussion Thread

(October 20th, 2016, 13:09)rho21 Wrote: Ah, it's available at last. I see they've cut the bit where Christopher launches his baton into the crowd (see that he has it before the tubular bells close-up section at the end, and doesn't have it afterwards).  lol

lol

Was it intentional?
Reply

(October 23rd, 2016, 01:26)ipecac Wrote:
(October 20th, 2016, 13:09)rho21 Wrote: Ah, it's available at last. I see they've cut the bit where Christopher launches his baton into the crowd (see that he has it before the tubular bells close-up section at the end, and doesn't have it afterwards).  lol

lol

Was it intentional?

No, he was just really excited and waving his hands around too hard as the piece reached its climax. lol
It didn't do any harm, thankfully. The concert was really good - Ruff_Hi, I hope you enjoy the one in New York.
Reply

(October 22nd, 2016, 23:30)Quagma Blast Wrote: The hidden agendas are random... but Gandhi is specially defined to have a somewhat higher than normal chance to have Nuke Happy as his hidden agenda. (Teddy Roosevelt also has a preferred hidden agenda, in his case Environmentalist to play nicely with his National Park-boosting ability.)

They're supposed to have "checks" to make sure the hidden agendas don't conflict in some way with their primary agenda, but I'm not sure if they all work. I think I saw a screenshot of Cleopatra with the Paranoid hidden agenda, which states she dislikes Civs that have a strong military... which is the exact opposite of her primary agenda.

But yeah, Teddy definitely seems to have a strong bias towards the Environmentalist agenda. I've seen him have it in like every game so far.
Civ 6 Adventure 1 Report
Now complete!
Reply

Yeah, they're definitely not fixed, but probably not 100% random. This is the first game I saw Gandhi in, so can't say how often he shows up nuke happy smile Teddy does show up as Environmentalist way too often, perhaps there's a bug which makes his hidden agenda fixed? Needs more testing!

Have to agree on AI's inability to expand, but I suspect this is because the devs didn't know themselves what the optimal expansion path would be. The biggest flaw is that AI wanders around with the Settler for 20-30 turns before deciding what to do with it. And the direct consequence is their inability to keep up with the player in mid-game and falling hopelessly behind in the late game

A couple of screenshots from the game on King I finished last night

Here's Japan, the largest civ after myself. Not shown is Valetta, a city state to the north they captured late. So 9 cities, not too bad considering I am on 12. But only 3 are past size 10




On the flip side, here's Kongo. I expanded in their face, they eventually attacked me, lost the war and their capital.. so here they are, on two cities, with a crapton of obsolete units, sitting there and not trying to do anything. Perhaps AI shouldn't be eager to make peace when it's clear it's not going to recover?




Also, I can attest to tech speed being completely ridiculous in the late game. 2 turns for a final spaceship tech? 3 turns for Future Tech? shakehead
Reply

(October 23rd, 2016, 04:16)yuris125 Wrote: Have to agree on AI's inability to expand, but I suspect this is because the devs didn't know themselves what the optimal expansion path would be.

Civ4 worked out its core gameplay with more than a year left to development time. Soren's talked about it in various discussion panels.

At the time, I thought this pattern was normal for game development. (It worked so well!) Just goes to show how wrong you can be on too small of a data sample.


AIs don't need an optimal expansion path. (If there is such a thing, game balance needed more work before AI work even began). They do need to be competitive on city count, though. This is something known since Civ2 days, and fixed (for Civ) since Civ3.

If the AIs are indeed underwhelming on expansion, that should be a top priority balance fix, right behind anything that crashes the game. How could this problem even be allowed to creep back in to the franchise? I suspect that Civ5's minimalist city counts may have set up an environment in which the AIs didn't need to expand properly any more. If the tight grip on useful city count was loosened, but the Civ6 AI is built on the Civ5 AI... You can see where that could lead.

Another potential explanation is that expansion has been loosened but there is too narrow of a needle to have to thread to pull it off, and the AI just can't do it. If that's the case, then lack of simplicity in design is biting them, and this would be much harder to fix than the AI simply not giving expansion enough priority.


- Sirian
Fortune favors the bold.
Reply

(October 23rd, 2016, 07:24)Sirian Wrote: Civ4 worked out its core gameplay with more than a year left to development time. Soren's talked about it in various discussion panels.

At the time, I thought this pattern was normal for game development. (It worked so well!) Just goes to show how wrong you can be on too small of a data sample.

As a professional games developer (sadly not on anything strategy),  lol  lol  rolf. How much I wish I could have worked on Civ IV. smile

I once worked on a game where a large proportion of the design changed completely 7 times in the final six months before launch. That was not much appreciated by the development team.

Interesting speculation on the Civ VI AI. It seems that expansion may be more expensive than previously; I wonder if (in many cases) the AI is expanding at the correct rate for its level of economic development, it's just dropped too far behind in that department.

I'm loving what I'm reading about the need to defend against barbarians early on. Sounds like a real stake in the heart of the farmer's gambit that's been such a prevalent feature of the series. Can only be a good thing in my book.
Reply

Ive now played two prince-level games where I got into early war (t 70-80 ish). 2 archers +warrior = win. It seems that enemy hardly counterattacks, perhaps because after a volley of arrows their expected result is pretty much defeat. After slaughtering their standing/mobile army, two archers plus a warrior took city easily. Perhaps because they had one promo, and I rocked with oligarchy, and they had no walls. Perhaps.

Anyway, both times I end up with a nice city, a defeated player, safe backlines, free settler. 

Bottom line, taking an early capital is easy and extremely profitable on prince. Might be less easy on higher difficulties, but I could have invested a lot more into the war, slightly hampering my cities, but I suspect the reward still means you come out ahead.

Unless AIs are gonna remember my BC war into modern times, I fail to see the problem, but I cant think of an easy way to nerf the 'early capital rush' strategy.
Played: FFH PBEM XXVI (Rhoanna) FFH PBEM XXV (Shekinah) FFH PBEM XXX (Flauros) Pitboss 11 (Kublai Rome)
Playing:Pitboss 18 (Ghengis Portugal) PBEM 60 - AI start (Napoleon Inca)
Reply

Here are a few more screenshots from the same game to illustrate AI's expansion issues

The peninsula to the north of America




Very reasonable land, with strategic and luxury resources. One problem is, it lies beyond a patch of desert. Was AI too scared to cross it?

America only had 4 cities by the end of the game, cramming the 4th in between India and Buenos Aires




The other area is to the south-east of France. I actually thought about setting it myself, but by the time I could do it safely, I was fast approaching the end game and was more focused on making sure I had as much production as I could get for space projects




Again, there are patches of tundra and desert, but very decent land in between, even with a natural wonder. Instead France crammed 3 cities very close together and didn't go beyond that

France had 5 cities at the end, with one more captured by Germany




All in all, AI seems to be excessively concerned about lushness of the land. You can settle cities with only good tiles early on, but mid-game, just cram them wherever you can! Especially in Civ6, where you can even make use of low-yield tiles by putting districts on them

Add to this the problem of AI not willing to declare war when warmonger penalties become too high (Kongo had higher military rating than me, literally fielding Spearmen vs Tanks and Crossbows vs Modern Armors. Come on man, do something with those units! They're crippling you already weak economy otherwise!), and we end up with AIs who can't keep up with the player in midgame and end up horribly behind by late game

I also feel that AI's logic for using Builders is flawed. Devs said they rewrote AI from ground up, but I can't imagine them not reusing at least some of the code. I wouldn't be surprised if the logic for Builders is just the old logic for Workers given a face-lift. Which just doesn't work, most notably because you have to keep building Builders throughout the game. AI ends up with a lot of unimproved tiles, and I think it's a direct consequence

That being said, Civ4 AI, especially pre-BTS, couldn't handle Workers particularly well either

One thing of note is that yesterday's Marbozir's video shows an AI with 6 cities on T100. That's not bad at all, and that's on Emperor difficulty. Emperor is the first level where AI starts with 2 Settlers, maybe that's enough of a kick-start to help with these problems?
Reply

(October 23rd, 2016, 08:13)Molach Wrote: Unless AIs are gonna remember my BC war into modern times, I fail to see the problem, but I cant think of an easy way to nerf the 'early capital rush' strategy.

Axe rush is a time-honoured strategy which was effective in Civ4 smile It is even better in Civ6 because of city states. Unfortunately I agree there's no real way to address the issue. I think Civ6 did their best by letting the AI attack city states as well (they will do that without much hesitation), and attacking a player is much less effective on high difficulties
Reply

Looks like first major "bug" has been confirmed.

Multiple copies of amenities don't increase happiness in additional cities. I really hope this is confirmed as a bug and the devs don't turn around and say this was intended.
BTS games: PB1, PB3, PBEM2, PBEM4, PBEM5B, PBEM50. RB mod games: PBEM16, PBEM20, PB5, PB15, PB26, PB27, PB37, PB42, PB46, PB71. FFH games: PBEMVII, PBEMXII. Games ded lurked: PBEM17, PB16, PB18
Reply



Forum Jump: