Putting the 'role' back in role-playing games since 2002.
Donate to Codex
Good Old Games
  • Welcome to rpgcodex.net, a site dedicated to discussing computer based role-playing games in a free and open fashion. We're less strict than other forums, but please refer to the rules.

    "This message is awaiting moderator approval": All new users must pass through our moderation queue before they will be able to post normally. Until your account has "passed" your posts will only be visible to yourself (and moderators) until they are approved. Give us a week to get around to approving / deleting / ignoring your mundane opinion on crap before hassling us about it. Once you have passed the moderation period (think of it as a test), you will be able to post normally, just like all the other retards.

Grand Strategy Crusader Kings III

Vatnik Wumao
Joined
Oct 2, 2018
Messages
19,617
Hew Dev Diary about performance & AI improvements.

Full threading CK3 on "a nice modern Ryzen 9 3900X, and my 16 GB of 2400 MHz RAM with 32 GB of 3600 MHz RAM":

:deathclaw:
That is retarded test,what is the point testing it on top tier machine? If it was mid/low level,it would have been nice,but nobody expect to run like shit on top tier machines lol.

Got that covered as well:

"Let's take a look at the difference threading makes. I set up a simple test; I ran the game for 1 minute on my machine at max speed, first with threading fully enabled, then with threading disabled. You can see it for yourself below. Left is with threading, right is without:

The red line you see in the video is the time between each frame. For 60 FPS it should be at or below the green horizontal line.

As you see, the difference is staggering. Without threading the framerate is dreadful, and the game progresses far more slowly. With threading the game progressed 958 days, while without it only progressed 546 days. That is, it ran 75% faster with threading, and with a far far better framerate.

The machine I’m running this on is my home PC, which at the time of recording had an i7 4770K, a GTX 1080, and 16 GB of 2400 MHz RAM, running at the highest graphical settings. The CPU and RAM are both quite old and by this point far outperformed by newer models, though the GPU is still solid. Max speed on this is high enough that I almost never use it in normal play, instead mostly using speed 3 and 4."
 

fantadomat

Arcane
Edgy Vatnik Wumao
Joined
Jun 2, 2017
Messages
37,558
Location
Bulgaria
:deathclaw:
Ahhh on the left you run it like 10fps lol,on the right you have too much frame fluctuation/staggering. The game could do with an option for fixed 30 fps.
 
Joined
Jan 7, 2012
Messages
15,272
Comparing fully threaded (which on his system would be up to 24) vs. no threads is kind of silly, do it with 4 or 8 threads like I assume 95% of people playing the game would have.

Still looks reasonably quick for only 1 thread though. Hopefully it stays quick after a few hundred years.
 
Joined
May 11, 2007
Messages
1,854,434
Location
Belém do Pará, Império do Brasil
The machine I’m running this on is my home PC, which at the time of recording had an i7 4770K, a GTX 1080, and 16 GB of 2400 MHz RAM, running at the highest graphical settings.

That's supposed to impress me? That's still a pretty powerful machine.
I swear, sometimes you gringos act like you can find 16 gigs of RAM on a random cupboard or something.

I want to see how it runs on a minimum requirements machine:

  • CPU: Intel Core i5-3470T 2.8GHz Or AMD FX-6100
  • Graphics: Nvidia GeForce GTX 650 Ti or AMD Radeon HD 7770 1GB GDDR5
  • RAM: 6 GB
  • Windows: Windows 7,8,8.1 and 10 (64 Bit)
  • DirectX: DX 11
  • HDD Space: 20 GB

If your game can't run well at minimum requirements with everything set on low, optimize it harder.
 
Joined
Jan 7, 2012
Messages
15,272
Single core performance has been pretty much stagnant for years, anyone with a dual core or quad core system should theoretically have much better performance than a new top of the line CPU that is artificially limited to 1 thread.
 

fantadomat

Arcane
Edgy Vatnik Wumao
Joined
Jun 2, 2017
Messages
37,558
Location
Bulgaria
That is retarded test,what is the point testing it on top tier machine? If it was mid/low level,it would have been nice,but nobody expect to run like shit on top tier machines lol.

And then I remember all the threads about dudes with Super Space NASA PCs complaining about how much of a badly-optimized piece of shit Stellaris is.
Yeah,i don't take such idiots seriously. It doesn't matter how good your pc is if you are a retard that fucks it up with its software. I have gtx580 and still run most shit fine,paradox game run pretty fast on my pc as an example. But i also don't clug my process bar with retarded apps and those new nvidia spyware and run practical old drivers.
 

thesecret1

Arcane
Joined
Jun 30, 2019
Messages
6,716
Single core performance has been pretty much stagnant for years, anyone with a dual core or quad core system should theoretically have much better performance than a new top of the line CPU that is artificially limited to 1 thread.
Yes, theoretically. In practice, many games do not support multiple cores, or if they do, they often only support two or four at most. I don't have much faith in Paradox doing well with performance in this regard (I mean, it will probably be fine on release, when they push out a hollow shell of a game with next to nothing inside, but after a couple years of content bloat, who knows?). Speaking in general, it is better to seek a balance between the number of cores and individual core performance
 

thesecret1

Arcane
Joined
Jun 30, 2019
Messages
6,716
I bet they'll add China in one of the DLCs. The way they spread the map all the way east, including Siberia, is clearly pointing at it. They just want to do it in a DLC so that they can focus on it some more (because they cannot copy paste any of it from CK2). Japan DLC will naturally follow, because if you already went and added China, why wouldn't you add Japan?

This content bloat tires me. Then again, once they cover the entire planet, including fucking America and Australia (muh abos), maybe they'll finally be forced to provide at least a smidge of flavour? Just a bit?
Who am I kidding? They'll just start adding bookmarks all the way to the stone age instead.
 
Joined
Jan 7, 2012
Messages
15,272
Single core performance has been pretty much stagnant for years, anyone with a dual core or quad core system should theoretically have much better performance than a new top of the line CPU that is artificially limited to 1 thread.
Yes, theoretically. In practice, many games do not support multiple cores, or if they do, they often only support two or four at most. I don't have much faith in Paradox doing well with performance in this regard (I mean, it will probably be fine on release, when they push out a hollow shell of a game with next to nothing inside, but after a couple years of content bloat, who knows?). Speaking in general, it is better to seek a balance between the number of cores and individual core performance

Yeah, but the fact that they provided a video comparing 1 thread vs 24 threads shows that the game clearly takes advantage of multithreading decently well.
 
Joined
Oct 1, 2018
Messages
2,323
Location
Illinois
That looked promising, though it's almost impossible to tell with Paradox games until you're actually playing them yourself. One area that was a noticeable change that I'm excited about, knights. Making them fewer super-units with actual characters is a huge plus in my book. Reminds me of Medieval Total War 1 where generals could get massively overpowered godlike stats and become obscene heroes by themselves. Adds more fluff and having fun if you get some mega-knight, or possibly if you go into a heavy martial character you can become a beast.
 

Zariusz

Liturgist
Joined
Nov 13, 2019
Messages
2,059
Location
Civitas Schinesghe
That looked promising, though it's almost impossible to tell with Paradox games until you're actually playing them yourself. One area that was a noticeable change that I'm excited about, knights. Making them fewer super-units with actual characters is a huge plus in my book. Reminds me of Medieval Total War 1 where generals could get massively overpowered godlike stats and become obscene heroes by themselves. Adds more fluff and having fun if you get some mega-knight, or possibly if you go into a heavy martial character you can become a beast.
I except that atleast few will probably try to make knights only army.
 

As an Amazon Associate, rpgcodex.net earns from qualifying purchases.
Back
Top Bottom