• We have updated our Community Code of Conduct. Please read through the new rules for the forum that are an integral part of Paradox Interactive’s User Agreement.

CK3 Dev Diary #103: Summer Break

Greetings!

Summer is upon us, and with Fate of Iberia and the 1.6.1 update(s) (note: another small fix update will be released this week), we feel confident that the game is in a good place where it can be enjoyed over the break! (As always, keep those reports up if you find an issue) Soon we’ll all be out enjoying the beautiful Swedish summer, taking some well-deserved rest, and recharging for the time ahead!

As mentioned in Dev Diary 100, we have big and small plans, some of which will be revealed later this year. We’re very excited to see what you think, but we can’t say anything yet!

From this point onwards we will not have any regularly scheduled Dev Diaries until we’re all back! There may be a small dev diary here and there, but no promises. We’ll be back before the end of August.

If you want to keep discussing the game over the break, head over to our Discord. Also, note that we’ll have videos every Wednesday and Friday throughout the summer, mostly of the DevClash that wrapped up recently.

Before we leave you for the summer, did you know that we added several historical figures that can appear under the right circumstances? Here’s an overview written by Ola, known as @Vaniljkaka here on the forums!

Abbas ibn Firnas (~810-887), known in Latin as Armen Firman, was an Andalusian polymath. Sources claim that he made an attempt at flight by jumping from a tower in Córdoba clad in a loose cloak stiffened with wooden struts. For this, he got immortalized, and now has a statue outside Baghdad’s airport. Though Ibn Firnas would be alive in our 867 bookmark, you’ll only encounter him if the right event fires.

al-Zahrawi (936–1013), known as Albucasis in Latin, was one of the greatest surgeons of the Middle Ages, and court physician to the caliph in Córdoba. If you manage to encounter him in the game (you’ll need some luck for this, and an event about medicine…), he’ll even tell you an anecdote from his career.

Ibn al-Wafid (997 – 1074), known as Abenguefith in Latin, was a pharmacist and agronomist in Toledo in the middle of the 11th century. Historical sources claim that he was a pupil of al-Zahrawi, but this seems improbable, as al-Zahrawi died before 1013. Ibn al-Wafid is a great court physician if you play emir Yahya in the Rags to Riches bookmark.

Al-Zarkali (1029-1100), known in Latin as Arzachel, was a craftsman and astronomer based in Toledo. He fled the city when Castille conquered it in 1085. His work formed part of the basis for the Toledan Tables, a set of astronomical tables that were later translated into Latin by Gerard of Cremona. He is in the court of Toledo in the 1066 start.

Joseph ibn Nagrela (1035-1066), also known as Joseph ha-Nagid, was vizier to the incompetent, alcoholic emir Badis of Granada. Ibn Nagrela belonged to a prominent lineage of Sephardic Jews; his father Samuel was a famous scholar, warrior, and poet. Ibn Nagrela was the chief victim of the Granada Massacre in December 1066, crucified by an angry mob for supposedly trying to usurp the throne. However, our game begins in January 1066, so perhaps he will fare better in this history…

Ibn Zuhr (1094-1162), known in Latin as Avenzoar, was a physician and poet, who seems to have had a dramatic fallout with one of his employers, the Almoravid ruler. This fallout was very much the inspiration for one of the events in Struggle for Iberia…

Ibn Tufail (1105 – 1185), known in Latin as Abubacer Aben Tofail, was a physician, novelist, and astronomer, among other things. He was quite keen on autopsies, as you’ll notice if you encounter him, which will require the right event at the right time.

Gerard of Cremona (~1114-1187) was an Italian translator of scientific books from Arabic into Latin, active in the kingdom of Castille. After the fall of Toledo, his work was instrumental in making Arabic knowledge available to the Western European intellectual sphere, ushering in the “Renaissance of the Twelfth Century”. Gerard can be encountered if you get an appropriate event during the years when he was active - make sure that conciliation is the prevailing mood in Iberia!

Ibn Rushd (1126-1198), known in Latin as Averroes, was a polymath and jurist and one of the most influential intellectuals of the Middle Ages, with a whole school of thought, averroism, that bears his name. In his youth, he seems to have been a pupil of both Ibn Zuhr and Ibn Tufail. There seem to be claims that he experimented with flight, just like ibn Firnas. Though he is very famous, he lived after our game’s current bookmark dates, and you’ll only encounter him with a bit of luck through an event in the right time period…

Qasmuna (11th-12th century) was a female Sephardi poet. Some of her poems are preserved, but little is known of her life. However, there is a theory that she was the sister of Joseph ibn Nagrela. I chose to go with this, since it made their family tree more interesting and allowed us to include her in the game. You’ll likely find her with her brother in the court of Granada.

We are excited to come back refreshed and relaxed after vacation, and resume working on the game that we all know and love! Until August, goodbye!
 
  • 72Like
  • 24
  • 17
  • 7Love
  • 5
  • 3Haha
Reactions:
I need HARD difficulty,the game is too easy for me now.

And i also hope hard difficulty is not only buffing the ai,but also make ai more clever.
 
Last edited:
  • 2Like
  • 2
  • 1
Reactions:
I need HARD difficulty,the game is too easy for me now.

And i also hope hard difficulty is not only buffing the ai,but also make ai more clever.
AI is a very difficult thing for grand strategy games, if you don't just buff them.
Computational power you can allocate for that needs to be minimal per actor in a game like CK3 with many actors and the realm of possibilities is incredibly vast.
Long term planning is a major advantage that humans have over simple AIs. Simple AIs usually evaluate actions on immediate benefit (aided by preset weights given by developers), because any long term benefit is more difficult to bring into play. Simple AIs don't have long term goals, just opportunities and limited means to evaluate these.

An additional difficulty is that Paradox obviously wants different personalities to act differently.
So creating many different AIs that all play well is even more of a problem.
To make things even worse, DLCs and updates widen the realm of possibilities immensely, not just by adding more options, but also by introducing various combinations of DLCs enabled. Maybe the AI is really good at utilising mechanic X, but then someone plays with all DLC but the one that includes X and then it spirals out of control, because the devs didn't extensively playtest every possible combination of DLCs properly to see if the AI relies to heavily on certain mechanics.

In general, i agree with the sentiment, but i think success is going to be limited, because of the huge dev time investment that would be necessary to make the AI actually good at the game. I think we'll have to settle for acumulating minor improvements here and there.
 
  • 2Like
  • 2
  • 1
Reactions:
AI is a very difficult thing for grand strategy games, if you don't just buff them.
Computational power you can allocate for that needs to be minimal per actor in a game like CK3 with many actors and the realm of possibilities is incredibly vast.
Long term planning is a major advantage that humans have over simple AIs. Simple AIs usually evaluate actions on immediate benefit (aided by preset weights given by developers), because any long term benefit is more difficult to bring into play. Simple AIs don't have long term goals, just opportunities and limited means to evaluate these.

An additional difficulty is that Paradox obviously wants different personalities to act differently.
So creating many different AIs that all play well is even more of a problem.
To make things even worse, DLCs and updates widen the realm of possibilities immensely, not just by adding more options, but also by introducing various combinations of DLCs enabled. Maybe the AI is really good at utilising mechanic X, but then someone plays with all DLC but the one that includes X and then it spirals out of control, because the devs didn't extensively playtest every possible combination of DLCs properly to see if the AI relies to heavily on certain mechanics.

In general, i agree with the sentiment, but i think success is going to be limited, because of the huge dev time investment that would be necessary to make the AI actually good at the game. I think we'll have to settle for acumulating minor improvements here and there.
I don't think that's the whole story.

Coding in general is hard, not just AI. Also, you don't need to buff every character AI, just those who are engaging with the player. A dumb AI doesn't need to respond smartly to a move done by other dumb AI.

I think there is not much incentive for the developers to make a better AI. First of all, most players are casuals who don't want to be beat by the AI and second, buffing the AI is easier and most players don't notice the difference.

It is a subtle difference to what you said, but I think it is relevant to point out.
 
  • 1Like
  • 1
Reactions:
There are mods that significantly improve the AI just by tweaking some values, so there's some low hanging fruit that doesn't require difficult coding.

One reason Paradox has occasionally given for it is that if the AI becomes good at developing its provinces, it ends up making the player more powerful (the player conquers better-developed provinces)
 
One reason Paradox has occasionally given for it is that if the AI becomes good at developing its provinces, it ends up making the player more powerful (the player conquers better-developed provinces)
Yeah, that was also the reasoning behind AIs losing their buffs in stellaris when vassalized. But it led to economy death spirals, because all of the sudden all buffs to job output got lost and you didn't have enough basic goods for the higher goods, which caused all types of deficit debuffs.
They have revised that with the last DLC Overlord (Thematically about vassalisation) to reduce the buffs instead of removing them - accompanied by improving AI economy management and how deficits work. So the AI economy is in a lot more stable place right now and doesn't death spiral indefinitely anymore.
 
  • 2Like
Reactions:
One reason Paradox has occasionally given for it is that if the AI becomes good at developing its provinces, it ends up making the player more powerful (the player conquers better-developed provinces)
this is also the reason given for why the eu4 team refuses to make their ai diplo-dev and save a slot for manufactories. galaxy brained takes all around

the thing about the province just being better for the player when they eventually take it is it also makes the province better for the ai while they hold it, which makes it harder for the player to take it. you know, because the ai has a better realm, and are stronger, and have better economies. i feel like we shouldnt have to explain this
 
  • 1Like
  • 1
Reactions:
  • 3Haha
Reactions:
For a bit of context, the everyone on the Paradox team probably* don't have two months vacation right now.
That's just the vacation-period for the team as a whole, where they can't keep doing all the normal workflow due to a lot of people being on holiday.

The individual people on the team will likely have a good chunk of that off, but not all of it, and probably offset a bit from each other so that the company can keep doing those things that really can't wait (like paying their bills and so on).
Plus a few low intensity things that can be done without needing the entire team, but still fewer of those than they can normally do when everybody is in.
For instance those video's they are going to be uploading during the summer won't make themselves.

Now don't get me wrong they'll still likely have significantly more vacation time both during the summer and during the year, just probably not 2 full months in a row for the entire team.
(under Swedish law the absolute minim people are entitled too is 5 weeks vacation a year + holidays and a right to have 4 weeks in a row during the summer months if they want it. AIUI it's more common than not for people's employment contracts to give them a bit more than the legal minimum. Benefit of having a bunch of strong unions in society who influence what's considered "normal" for a job to offer.)

*I don't actually know the specific situation at Paradox, I just have a decent understanding of the general situation in Sweden.
Your understanding of the law is correct. :) It's quite rare to have more than the five week-minimum, though. That's really only a common thing for older people in the public sector, so most Paradox employees likely "only" have five week vacations.
 
  • 2
Reactions:
How long is the Summer holiday in Sweden?
They come back late November, then work a couple of weeks before breaking up for Christmas - which in Sweden lasts until the extended spring break in March. On average people work around twelve weeks a year, and that's mostly from home.
 
  • 6Haha
  • 2
Reactions:
You guys should be working. Every month that goes by where you're not developing the next DLC, that's days that a competitor can use to develop a game just like Crusader Kings and do it better.
Paradox was on the other end of this scenario less than 10 years ago with Sim City. They had Cities: Skylines waiting in the wings when Sim City stumbled, and EA lost a lot of revenue to Paradox.

Now I'm just waiting for the other shoe to drop with Crusader Kings. It's been >>>THREE YEARS<<< and we basically have 2x DLCs to show for it: Royal Court, and this Iberia struggle stuff. (Don't kid yourself in thinking the Northern Lords flavor pack counts toward this. Every development team needs to have some features already spooled up prior to release, and I find it hard to believe NL wasn't one of them.)

Politically speaking, I guess you could say I'm a socialist in the American realm, which basically equivocates me to being a centrist from a European perspective: time off is good (especially if you just had a kid), and healthcare should be nationalized, not employer-run. But this is too much. You're opening yourself up to exploitation by a competitor the longer you take on this stuff.
 
  • 5
  • 3Haha
  • 2
Reactions:
Pardon me, >>>2 years and 10 months<<<, or maybe we can say almost 4 years since the first dev diary. If I'm Maxis or Ensemble, I'm looking to break into this field as soon as PDX is caught sleepin...which they are, right now!

Put another way, looking at the real estate of the whole map, it's pretty funny--in a bad way--that 1/8 of the game is not worth playing due to the ERE's inheritance law currently reading as "Primogeniture". You could maybe even say 1/3 of the game isn't worth playing if your goal is to recreate the roman empire from Byzantium.
 
  • 1
Reactions:
Pardon me, >>>2 years and 10 months<<<, or maybe we can say almost 4 years since the first dev diary. If I'm Maxis or Ensemble, I'm looking to break into this field as soon as PDX is caught sleepin...which they are, right now!

Put another way, looking at the real estate of the whole map, it's pretty funny--in a bad way--that 1/8 of the game is not worth playing due to the ERE's inheritance law currently reading as "Primogeniture". You could maybe even say 1/3 of the game isn't worth playing if your goal is to recreate the roman empire from Byzantium.
1 year and 10 months. It was released in September, 2020.
 
  • 4
  • 1Haha
  • 1
Reactions:
Nice summer break guys, I wish I can have this long break at my employer <3