Performance Megathread

[Foxosaur]

Why bother with performance if you can just churn one DLC after another? The game is already like 4 years old. There's no economic gain in allocating resources to deal with persistent issues. It won't bring any new customers and old suckers customers who already bought it have gone into the sunken cost fallacy so they can be easily herded back with promises of minor improvements.

Hello,

I am sure for long term people of the community, it may seem that Stellaris doesn't get new players but it does, I only brought the game (DLCless) in April and I am disappointed to learn and more shocked the more I have read, the bad (lack?) handling of general performance problems. I have since brought 5 DLC(I think, I cant remember), played nearly 240 hours and I'm hesitant to buy more, yet really would like too. And aside taking my perspective as completely unique or my being new completely unheard off, I'm in discord a fair amount to know there is regular brand new players that come.

It would be wholly ignorant of team managers and none-technical people to believe that there is no revenue from keeping a community healthy and happy with their content (so they are more likely to then hang around and buy more). I would expect that if they did commit to sometime to unravelling their game and reorganising it, they could see long term gains and retaining of playerbase by 20% or something very favourable for every DLC they release. Its just unfortunate that the reality must either be that level of ignorance or their code base is such a mess (maybe high churn of team members) that its very difficult to unravel.

Anyone with a good sense of community / business perspective would acknowledge that there is value in pumping time into it. Otherwise the cost will remain blindly/naively high but 'invisible' if you simply choose not to look at it.

 

[GnoSIS]

It's literally incredible that they were eager to announce a decrease in loading screen times. I've never seen a company maybe outside of EA that has ever been so fundamentally incompetent and disconnected from their consumer base. And everytime they announce that nothing about performance will be fixed at all the massive horde of die hard brand-loyalists come out in full swing defending everything paradox does. This company could sell them a plate of dog shit and they would ask for more. Not to mention the almost complete ignoring of this thread by the Devs AND the low-key unannounced censorship that goes on in these forums make for one hell of a negative customer experience. And im sure this post will get censored too.

You don't get it, this thread is the graveyard/censorship for comments regarding performance. Beacuse who is going to read a comment that has lost all it's context (no post title) on page 42 here?

The negative customer experience, goes on on the forums and inside the game as well. And as far as the super secret project of stellaris 2 is concerned, thanks but no thanks. With this kind of behaviour from this company I won't even buy Vicky 3 if it comes out.

 

[Paladinn]

Well the title is kinda clickbait, but this solution would absolutely work. In a nutshell:
Problem A: The game treats AI as it does for the player, making the math for each planet, each, pop, tradition, etc... You can imagine that in a large galaxy this is insane, maybe in 10-20 years the CPUs will be able to process that amount of data without lag. I think you can optimize as much as you want, but we will always have late game lag and I believe this is the reason the AI sucks so much in everything, devs can't improve it because it would suck even more resources from the CPU.
Problem B: The AI can't handle jobs distribution and ends up with bad economy.

Solution: People wouldn't like this solution and I'll get a lot of disapproval for this post, but this wont change the reality of things. If you just simulate AIs pops and economy, you fix all the problems. How do you do that? All AIs start like they do now, but instead of getting resources and stuff from pops, they get flat incomes calculated at the start, with some random factors. They would expand and build normally, so the player wouldn't notice any difference. When you click on AIs planets, you see pops and buildings normally, the difference is that the AI would ignore those things when calculating the income, the only things that matter would be:

• Empire's size in terms of pops and planets.
• Initial factors (empire civics, empire behavior, etc...).
• Random factors
• Difficulty

This way AI's economy would not be affected by bad jobs distribution.

However, people would not like this system because it feels like AI is cheating, but in the end you can't have late game performance and 100% simulated economy realism.

This one is quite good. If devs find a way to balance this I am okay with the "cheating" Performance of single player and ESPECIALLY multiplayer are the major dislikes I have towards this game. Also, we like it or not AI should cheat. They are not compatible. Even in grand admiral difficulty with scaling off and all advanced start, I simply keep up with all of them at 2270s 2280s. In my opinion, they should be at least comparable to 5x crisis strength when they are all at these settings when time is something like 2360s 2370s.

 

[GnoSIS]

Hello,

I am sure for long term people of the community, it may seem that Stellaris doesn't get new players but it does, I only brought the game (DLCless) in April and I am disappointed to learn and more shocked the more I have read, the bad (lack?) handling of general performance problems. I have since brought 5 DLC(I think, I cant remember), played nearly 240 hours and I'm hesitant to buy more, yet really would like too. And aside taking my perspective as completely unique or my being new completely unheard off, I'm in discord a fair amount to know there is regular brand new players that come.

It would be wholly ignorant of team managers and none-technical people to believe that there is no revenue from keeping a community healthy and happy with their content (so they are more likely to then hang around and buy more). I would expect that if they did commit to sometime to unravelling their game and reorganising it, they could see long term gains and retaining of playerbase by 20% or something very favourable for every DLC they release. Its just unfortunate that the reality must either be that level of ignorance or their code base is such a mess (maybe high churn of team members) that its very difficult to unravel.

Anyone with a good sense of community / business perspective would acknowledge that there is value in pumping time into it. Otherwise the cost will remain blindly/naively high but 'invisible' if you simply choose not to look at it.

Devs: "Let's milk Stellaris to the point of releasing Stellaris 2, and then go through the DLC treadmill again."
Can guarantee you, that I won't fall for this. S2 has to be an extraordinary game.

 

[alteration]

This one is quite good. If devs find a way to balance this I am okay with the "cheating" Performance of single player and ESPECIALLY multiplayer are the major dislikes I have towards this game. Also, we like it or not AI should cheat. They are not compatible. Even in grand admiral difficulty with scaling off and all advanced start, I simply keep up with all of them at 2270s 2280s. In my opinion, they should be at least comparable to 5x crisis strength when they are all at these settings when time is something like 2360s 2370s.

See your dislikes, people don't like that. I mean AI already cheats on higher difficulties (higher incomes), at this point make it full cheat and fix the game

Devs: "Let's milk Stellaris to the point of releasing Stellaris 2, and then go through the DLC treadmill again."
Can guarantee you, that I won't fall for this. S2 has to be an extraordinary game.

But you will fall for this, they will present it to you as the greatest game and a redemption for stellaris 1, in reality it will be "ah s**t here we go again"

 

[evdjj3j]

There are no monolithic ryzen cpus. Monolithic is a specific term that refers to having everything on a single die. No ryzen chip has this, since, at the very least, the IO is a separate die from the cores/cache.

Ryzen mobiles are monolithic, one big die. Look at the link I posted. You mark my post as wrong when you're the one who is wrong. The Ryzen mobile chips have always been monolithic. My link clearly shows that

 

[GnoSIS]

See your dislikes, people don't like that. I mean AI already cheats on higher difficulties (higher incomes), at this point make it full cheat and fix the game


But you will fall for this, they will present it to you as the greatest game and a redemption for stellaris 1, in reality it will be "ah s**t here we go again"

I guess time will tell, and my badges on the right <-

 

[James Fire]

Ryzen mobiles are monolithic, one big die. Look at the link I posted. You mark my post as wrong when you're the one who is wrong. The Ryzen mobile chips have always been monolithic. My link clearly shows that

Nope. The graphics are still not on the die.

 

[Mailosz]

Hi, I just want to add my story to the thread. I wasn't really into this whole Stellaris thing, i'm more of an EU fan. But i've got this almost free 3 months of xbox game pass, and I used it to look into Imperator (meh) and this game. And I was hooked! This game is really awesome, and I decided to buy it after my game pass ends.

But then I got into the late game. And it is totally unplayable on my (maybe not high tier, but quite ok) machine. So I won't buy your game even though I would really like to, because I can't really play it to the end. And that means that I also won't buy any DLCs.

So yes, you are actually losing potential playrs because of this issue.

I'm occassionaly checking this thread though to see if performance improved, my wallet is ready

 

[brantodb01]

So I got CKIII and decided to just hope and give it a go on this crappy laptop (Ryzen 5 2500U) I have an after turning the graphics down that game ran SO much faster then any paradox game I've played (besides CKII). So I'm wondering why Crusader Kings runs better on my crappy laptop then other PDX games (mostly Stellaris) do on my Ryzen 7 2700 Desktop

 

[Sindulfo]

Not a lot of technical expertise in these comments. People wondering why their Stellaris galaxy struggles in endgame when Doom Eternal can simulate a hitscan weapon and five whole enemies on screen without a hitch.

Between all the cathartic whining, there's some good convo from the more engineer-minded people trying to come up with actual solutions.

 

[trybald]

I'm occassionaly checking this thread though to see if performance improved, my wallet is ready

Lol, good luck with that!

 

[Exasperation]

So I got CKIII and decided to just hope and give it a go on this crappy laptop (Ryzen 5 2500U) I have an after turning the graphics down that game ran SO much faster then any paradox game I've played (besides CKII). So I'm wondering why Crusader Kings runs better on my crappy laptop then other PDX games (mostly Stellaris) do on my Ryzen 7 2700 Desktop

It's likely because they put a LOT of effort into optimizing during the lifespan of CK2 (and presumably CK3 has learned from CK2's struggles) - I remember when CK2 had the same sort of performance complaints as Stellaris does now (and for a similar reason - proliferation of characters, mostly unimportant ones, during the mid-to-lategame; all plotting away and consuming CPU cycles despite most of them not having any meaningful impact on the game).
I also remember one CK2 version in particular where the performance was particularly bad because of a bug where every Greek noble in the game would run through the entire worldwide list of male characters at regular intervals to consider which ones they wanted to castrate, regardless of how far out of reach those people were.

 

[GnoSIS]

So I got CKIII and decided to just hope and give it a go on this crappy laptop (Ryzen 5 2500U) I have an after turning the graphics down that game ran SO much faster then any paradox game I've played (besides CKII). So I'm wondering why Crusader Kings runs better on my crappy laptop then other PDX games (mostly Stellaris) do on my Ryzen 7 2700 Desktop

Of all the other reasons, a major one is the fact that it' doesn't have any DLCs yet to weight it down.

And guess what, Stellaris was the same on release!

If it was otherwise, management would come down and say "we can't use this as a vehicle for DLC profit! get back to work and clean it UP!"

 

[Rosthun]

Stellaris on the *actual* release was even worse than it was right after megacorp dropped, I forget the details but at first you couldn't hide the resources in each system in the galaxy map and after some time the game was literally unplayable as there was so many space stations the GPU couldn't render then all. Either it was simply icons on the galaxy map or the game tried to simultaneously render the inside of every discovered system. Makes you wonder how in the hell no one noticed that before the literal *release* of the game.

 

[GnoSIS]

Stellaris on the *actual* release was even worse than it was right after megacorp dropped, I forget the details but at first you couldn't hide the resources in each system in the galaxy map and after some time the game was literally unplayable as there was so many space stations the GPU couldn't render then all. Either it was simply icons on the galaxy map or the game tried to simultaneously render the inside of every discovered system. Makes you wonder how in the hell no one noticed that before the literal *release* of the game.

Those were release bugs and if you had apowerfull GPU you wouldn't even notice them. Besides they were quickly fixed more or less. But my reply that CK3 and any new PDX game lacks feature bloat, points to the notion that PDX will add it later piece by piece.

 

[evdjj3j]

Nope. The graphics are still not on the die.

Ryzen Mobile is 100% a monolithic die. If you would look at my links you would see that.

This one has an shot of the die.

Ryzen 5 2500U - AMD - WikiChip

Ryzen 5 2500U is a 64-bit quad-core mid-range performance x86 mobile microprocessor introduced by AMD in late 2017. This processor is based on AMD's Zen microarchitecture and is fabricated on a 14 nm process. The 2500U operates at a base frequency of 2 GHz with a TDP of 15 W and a Boost...

en.wikichip.org

The Anandtech article talks specifically about the die being monolithic.

With a monolithic design, AMD doesn’t need to apply such rigid standards to maintain performance. In Ryzen Mobile 4000, the Infinity Fabric remains on the silicon, and can slow down / ramp up as needed, boosting performance, decreasing latency, or saving power. The other side is that the silicon itself is bigger, which might be worse for frequency binning or yield, and so AMD took extra steps to help keep the die size small. AMD was keen to point out in its Tech Day for Ryzen Mobile that it did a lot of work in Physical Design, as well as collaborating with TSMC who actually manufactures the designs, in order to find a good balance between die size, frequency, and efficiency.

AMD’s Mobile Revival: Redefining the Notebook Business with the Ryzen 9 4900HS (A Review)

www.anandtech.com

I'm not sure why you want to continue to mislead people. Is your ego so fragile that you can't admit being wrong?

 

[Shrike]

Quick question. Or 2.

While I know, and am angry with the DLC and performance degradation (to the point that I haven't played an entire game since Megacorp).

The major problem is the lack of threaded optimisation, right?

So if I have a i7 9700k (new gaming laptop, haven't risked playing yet since being burnt in the past) with a higher boost clock such as 4ghz + for a couple of cores I will have a better chance of actually having a playable framerate?

Thanks

 

[UltimateTobi]

Quick question. Or 2.

While I know, and am angry with the DLC and performance degradation (to the point that I haven't played an entire game since Megacorp).

The major problem is the lack of threaded optimisation, right?

So if I have a i7 9700k (new gaming laptop, haven't risked playing yet since being burnt in the past) with a higher boost clock such as 4ghz + for a couple of cores I will have a better chance of actually having a playable framerate?

Thanks

Basically yes, the more GHz, the better. But with Stellaris, the better the rig, the more you delay the lag, it's inevitable.

 

[evdjj3j]

Quick question. Or 2.

While I know, and am angry with the DLC and performance degradation (to the point that I haven't played an entire game since Megacorp).

The major problem is the lack of threaded optimisation, right?

So if I have a i7 9700k (new gaming laptop, haven't risked playing yet since being burnt in the past) with a higher boost clock such as 4ghz + for a couple of cores I will have a better chance of actually having a playable framerate?

Thanks

I sent a message to Aspec who makes a lot of tutorials because I was wondering if my computer would be fast enough for the late game and he told me I shouldn't have any trouble with my 2700x. He said he's running a 4770 from 5 years ago and it works just fine for him.
Here's a link to his YouTube channel, he has great tutorials.

ASpec

Sci-Fi Central If you're curious on which mods I run by default, please check out the collection. There's nearly zero game mechanics impact, mostly visuals a...

www.youtube.com