Askia

[SuperNoobCamper]

I think the general consensus is Askia is one of the easiest civs to play with -even with MUC4VP making mandcav a Knight replacement- and a truely fearsome opponent which is something i had experienced in my last game as Austria; i managed to win a diplo victory before his hoardes of units could swap their attention to me .... He is just too good at doing pretty much everything and the AI seems to do very well with it which is kind of nice and scary at the same time.
The most common scenario is an overseas Askia wiping every civ within their reach and taking a couple of cities in a matter of a single turn.
I'm not sure if Tabya or Mandcav needs tweaking or it's just the AI is really competent at using it.
Bonus screenshots:

Spoiler I don't think i should get on his bad side :

Spoiler Manufactory ? who needs that anyway :

 

[azum4roll]

Maybe they're overperforming in this patch because of the extra amount of stone? Try the next version first.

 

[tothePAIN]

The changes to the tactical AI made Authority AI more dangerous. The AI can easily use Songhai's abilities, so Songhai is really strong on this patch. Even after Mancav obsolete, Songhai is still a dangerous warmonger because of their abilities with regards to rivers. Tanya's are simple to use and mean that Songhai has the infrastructure to go along with everything.

I'm currently playing a Zulu game where I'm #2 and Songhai is #1 and I think Songhai is showing why I dont think Zulus are overpowered. I cant sit my troops in his territory because they'll die instantly due to the anti-warmonger penalty.

 

[andersw]

So is it about time Songhai got a bit nerfed and if so how?
If no, what would be a reasonable nerf if I do my own modding?
I find it way to easy to steam roll early as Askia myself.
If I "see" undiscovered civs getting erased in classic I know Askia is on the other side and he wont stop unless I stop him, emperor AI is not close to be able to cope with him.
Also if YOU find Songhai cav is strong, the AI vs AI doesnt have the same warmonger penalties so he is basically playing vs cheiftain difficulty AIs.
Is he as extreme on immortal and diety as well?
(and yes I've had the occational game where he's been screwed but in general he explodes)

 

[Stalker0]

So is it about time Songhai got a bit nerfed and if so how?

For me, its not that Askia is OP at the moment, its that warmonger civs have finally gotten to the level of AI intelligence that they actually show up late game, as opposed to falling off and then its always the science runaways I'm dealing with. So I think its premature for a nerf, I definitely see Askia in top scores more than I once did, but compared to a Mayan, Ethiopia, or Morocco who I am often see in the top charts...I don't think they are top dog or anything.

Now as far as in human hands, I'll let others who have more experience with them comment.

 

[phantomaxl1207]

Being able to move alongside rivers, which are very oppressive on movement, as they are roads, is very powerful. This alone is huge change to how terrain affects warfare. And that's not all of the UA. Throw in extra production on an Ancient Era UB and you have a an early game monster. Not to mention an early UU horseman that can hit cities with the standard penalty.

 

[andersw]

For me, its not that Askia is OP at the moment, its that warmonger civs have finally gotten to the level of AI intelligence that they actually show up late game, as opposed to falling off and then its always the science runaways I'm dealing with. So I think its premature for a nerf, I definitely see Askia in top scores more than I once did, but compared to a Mayan, Ethiopia, or Morocco who I am often see in the top charts...I don't think they are top dog or anything.

Now as far as in human hands, I'll let others who have more experience with them comment.

Current game (emperor), and a common scenario with Askia on Pangea.
Rome had more or less non stop wars with Brazil and took one or two cities from start -> gunpowder.
This is along the line of expectation with a warmongering AI, some success but the AI takes some time, he should probably even have done a bit better.

In the same time Rome took 1-2 cities....

Askia:
ERASE Egypt (in ancient or early classic era), yes egypt can be a pushover.
ERASE Japan, a very turtly civ who should be able to hold.
Vassal China (have not taken capital).
Took 2-3 Brazil cities.
Conquered a couple of CS along the way.

This is far from the first time I see this happen and there is no way in hell any other civ would be close to do this.
Mongols and pre nerf Spain could snowball decent but not nearly this good.
I can see diety AI holding up a bit better vs Askia but Im not able to deal with that difficulty.

 

[azum4roll]

The Zulus can definitely do the same thing, just not as early. Most warmongers seem to struggle at cross-continent conquest against a (military) tech runaway though.

 

[andersw]

Managed to grind out a win after all and looked at the replay.
Epic Speed Askia:
Two city Egypt dead T113.
Brazil two cities taken T123 and T130.
siesta
Japan, Cities taken T225, T230, T240 and erased T249.
Two CS around T260-265 and vassal china.

 

[DeAnno]

The roads along rivers thing is really intense, it has a lot of knock on effects even later in the game. Specifically it allows you to move and attack through rough terrain in ways you wouldn't be able to, even in enemy territory where you wouldn't ordinarily be able to have roads. This routinely means an extra hit on a target unit and will often be the difference between it living and it dying. The AI usually has a little trouble moving units and attacking really efficiently, maybe this is allowing it to reach critical mass on offense more often and win battles with swarms of the horse UU? With AI muscle you really wouldn't need any kind of combined arms at all, a lot of civs would just fold to a swarm of Mandelekus probably even after Walls and contemporaneous units come down. They can also move around the empire in a flash and won't spend tons of turns marching back and forth between battles and tributes which maybe eases things on the AI strategically.

Beyond that the whole setup is very synergistic. You always go Authority and the AI will do that and the Mandelekus will then heal on kills and be even better and the Authority hammers and Tabiya hammers means the Songhai can outtrade the other AIs unit for unit and the Mandelekus can charge through the empire fast enough to overwhelm and make that matter maybe? I think on Deity the other AIs might produce fast enough anyway that the advantage is lost but even on Immortal I think it's probably still a big deal.

Like, overall, I think the military AI still excels at trading units above everything else (being safe and careful is so much harder), and Aksia is the king of trading.

 

[Mad Madigan]

I still support @pineappledan 's 3/4UC change to Askia being moved to vanilla VP: Mandekalu Cavalry being moved from Horsemen to Knight replacements in order to nerf their ability to easily crush Civs in the Ancient era.

 

[Stalker0]

I still support @pineappledan 's 3/4UC change to Askia being moved to vanilla VP: Mandekalu Cavalry being moved from Horsemen to Knight replacements in order to nerf their ability to easily crush Civs in the Ancient era.

Also from my very brief google searching, it seems Mandekalu were more common in the 13th century and were more associated with Mali than Songhai. So they wouldn't have been absorbed by Songhai until they were conquered by Mali in the 1300s or later when they ascended over Mali in the 1400s.

Aka...knights seem a more thematic timeframe for the unit, which supports your argument.

 

[pineappledan]

I think they were moved forward because there are already lots of knight replacements and G dropped the only unique horseman adding UBs to the 2 UU civs (Companion Cavalry). They were knights in vanilla because that was the most appropriate place for them, but I can see how an argument can be made for an earlier unit. It is an absence of evidence argument though.

The Mandinka warrior society existed before both the Mali and Songhai empires, but there isn't very good records from before that.

Since 4UC does add more horseman UCs, we didn't feel pressured to keep the Mandekalu there, and we reasoned that the unit's abilities were too potent for the ancient era.

 

[stormfallen]

They were also moved to swap with cataphracts, which no longer made any sense as a horseman replacement once they were ancient instead of classical.

I think moving them back to knight replacements makes the most sense, and maybe one of the other knight UUs should get pushed back to be a lancer replacement that unlocks a level earlier? Conquistadors would make sense as they were most notable during the 1500s, which matches with the early/mid renaissance. Possibly put them at Astronomy to best utilize their settling, and as a follow-up from Haciendas at Compass. Gunpowder or Chemistry would also work. Naresuan's Elephants would also make sense, with King Naresuan's famous elephant duel occurring in 1593, and his reign lasting between 1590-1605. They also have a similar niche to cataphracts as a less-mobile-but-beefier mounted melee.

 

[azum4roll]

But we'll still need a horseman UU. It feels strange without one, seeing as every other ancient era unit gets at least one.

 

[Melchizedek]

I play on Emperor and Askia hasn't been particularly strong in my games.

 

[stii]

I think it might be a case of at the higher levels the AI has a huge number of units and the extra movement benefits Askia a lot more there.

But it also might just be random chance, different patches have wildly different levels of AI aggression. So you'd need to play a huge number of games to really know. I have seen Mongols go crazy and conquer the whole world and they have a movement buff too, but who really knows. One game everyone has a DoF and the next everyone is at war.

 

[LifeOfBrian]

Honestly I'd be in favour of moving Mandekalu to Knight UU, because a horseman UU is just absurdly strong due to its mobility. I'd prefer if the non-modmod civs in VP didn't have an ancient era horseman UU. I can somewhat handle Askia on Deity if I start next to him, but 90% of the AIs can't, with the exception of civs with spearmen UU or the likes of Zulu/Swedes, so they usually balloon very early on in the game.

 

[Melchizedek]

I don't really mind if a civ has successful early conquest. That's what I'd expect to happen sometimes if I were playing humans. It's not unbalanced unless you almost define all winning as unbalanced. Seeing a civ that has early UU and took Authority conquer someone is exactly what should be expected in the game, it's not a balance problem. What I see much more often in my games, honestly, is civs taking authority and having early UUs and never even taking a city.

That said, Askia isn't even regularly conquering in my games or taking an early score lead, so I doubt there is anything wrong with them right now.

 

[LifeOfBrian]

Melchizedek, I haven't seen any other AI civ be as consistently great as the Songhai in my Deity games, they're far ahead of the pack as far as I'm concerned.