Stellaris - Paradox new sci-fi grand strategy game

[MadMaxHellfire]

(receive my rating as a positive one)

 

[oldmanpaco]

I literally had to go to Urban dictionary to find out what UwU means. Fuck you all for making me do that.

 

[Tytus]

I literally had to go to Urban dictionary to find out what UwU means. Fuck you all for making me do that.

UwU

 

[thesecret1]

Spoiler

I decided to get the same mod and it was a great decision. All the shitty references Paradox threw in now read like some kind of an anime plot :D

 

[Sarissofoi]

Is game still lag to hell and back past mid game?
And do AI can u8se game features or still not?

 

[Norfleet]

Yes, and no.

 

[Delterius]

Stellaris became slow because the community and the developers were tackling the wrong issues all this time. We assumed it was pops and even trade at one point. But it's actually jobs.

Even broken the way it is Stellaris is the one game that I can always have fun with. CK2 became so bloated with broken paths to success and EU4 remained too barebones for so long and only made itself cheaper with dumb power creep that I only play those games when I really have a want for their time period. Stellaris is the only thing close to Vicky and it's fully customizable.

But yes Federations is a bit of a meh. I'm looking forward to Origins and I guess the galactic community. Having psionic duels to decide the federation leader is cool and all.

 

[Preben]

Stellaris became slow because the community and the developers were tackling the wrong issues all this time. We assumed it was pops and even trade at one point. But it's actually jobs.

Even broken the way it is Stellaris is the one game that I can always have fun with. CK2 became so bloated with broken paths to success and EU4 remained too barebones for so long and only made itself cheaper with dumb power creep that I only play those games when I really have a want for their time period. Stellaris is the only thing close to Vicky and it's fully customizable.

But yes Federations is a bit of a meh. I'm looking forward to Origins and I guess the galactic community. Having psionic duels to decide the federation leader is cool and all.

The developers never actually tackled the issue. It was the community that deduced what was the problem, something that should be childishly easy with internal debugging tools.

Dunno what comes with the next patch, but so fat Paradox was largely unconcerned with performance and general quality issues. Their pipeline was entirely filled with churning out new features to cram into DLCs.

 

[thesecret1]

That's because you can't release a "performance improvement" DLC and milk people for it

 

[Norfleet]

Even broken the way it is Stellaris is the one game that I can always have fun with. CK2 became so bloated with broken paths to success

And Stellaris is BETTER? Stellaris is so broken that basically EVERY path leads to success because the AI is lobotomized.

Actually, the CK2 AI is also lobotomized, literally just picking actions at random with no plan or strategy at all. You just don't tend to notice because with so many actors in play, you get the million-monkeys approach, and "success" is somewhat less rigidly defined.

 

[Delterius]

And Stellaris is BETTER?

Not really.

 

[Delterius]

That's because you can't release a "performance improvement" DLC and milk people for it

Tbh their fanbase is rotten enough to buy a pure performance dlc.

 

[lophiaspis]

Reminder for anyone on the fence: This game is still irredeemable shit made by gormless pinheads. It sometimes appears to be improving, falsely, because the only members still regularly posting in this thread tend to be cretinous fanboys and similar bird-brained individuals.

 

[MadMaxHellfire]

please, show to the jurors where stellaris touched you by pointing on this doll.

 

[The Brazilian Slaughter]

Next dev diary is going to be about Performance.

I would say that this is going to be the Moment of Truth for this DLC and this game overall. Almost the entire fanbase is tired of being strung along by Paradox, with the exception of the fanboys.

I think they need a large performance upgrade to get back in the fans' grace. Say, at least +20% performance.

Even 20% is barely enough, I think this game would need to straight-out double its performance in order to actually become fully-functional.

Next dev diary is either going to make or break it.

 

[Sarissofoi]

Optimist.
New Dev Diary gonna be about some 'upgrades' that will tank performance even more.

 

[Quatlo]

Unless they ditch shitwitz engine there is nothing they can do. What were they thinking making this game on their crappy engine is beyond me.

 

[Preben]

Next dev diary is going to be about Performance.

I would say that this is going to be the Moment of Truth for this DLC and this game overall. Almost the entire fanbase is tired of being strung along by Paradox, with the exception of the fanboys.

I think they need a large performance upgrade to get back in the fans' grace. Say, at least +20% performance.

Even 20% is barely enough, I think this game would need to straight-out double its performance in order to actually become fully-functional.

Next dev diary is either going to make or break it.

Let's me honest. It's gonna be break it. They will announce some superficial optimizations that will show 15% "improvement" according to their internal testing. Completely irrelevant for player's experience if it takes several minutes to complete a single in-game year. In the end the game will work even slower because the newest patch/DLC will introduce features that will be even more taxing to the CPU such as Diplomatic Weight, which was already announced to be calculated by counting of all the pops and applying a shitton of modifiers.

The problem with Stellaris is that bad performance is caused not only by bad high-level programming, but also (and probably most importantly) by the aged engine. It's 2020 and the game is still unable to handle several cores properly. Just look at your CPU performance tab in Windows Task Manager while Stellaris is running. One core is used 100% most of the time, while other cores are mostly idle. Ironically, buying a newer PC is likely to make the game slower, because modern machines often use more cores rather than faster ones. But that's not all, in one thread about HoI IV the devs admitted that the engine part which governs the user interface is utter shit, badly designed and performance taxing, but they are afraid to touch it. The UI is so deeply embedded in the engine that working on it, let alone the necessary time and costs, could cause catastrophic cascade. And I guess that this applies to all engine-related problems.

 

[MadMaxHellfire]

what they were thinking? money.

 

[EmperorCruz]

I remember when buying games felt that you bought a complete experience, now its feels like your buying like 50% of a game and have to wait for a DLC to get more 10%.
Is what this game feels to me. Paying or 'performance dlc' is just too wrong. Another frustrate experience.

 

[Norfleet]

I don't think the performance is part of the DLC, that's just attached to the public patch that comes with the DLC.

The bigger point is that all of the DLC is actually already on your computer and so you're paying extra to use stuff you ALREADY HAVE.

 

[thesecret1]

It's 2020 and the game is still unable to handle several cores properly.

Sadly, that's not limited to Paradox. Many games are still incapable of using more than one, as are many engines. For example, IIRC, Unity only added multicore support in 2018. Hell, even games that do support multiple cores sometimes fuck up and put way more stress on one core than on the other ones. Always take this into account when buying a new machine.

 

[Average Manatee]

Its not really even a multithreading problem, the game is just badly coded. Its processing every frame shit that could be processed once a month with virtually no gameplay implications.

You can prove this, simply open the console and do ticks_per_turn x where x is 2-10. Everything about the game runs massively quicker, but because Paradox doesn't properly decouple the UI interactions from the rest of the game engine the UI becomes laggy and unresponsive. And this isn't even the limit of what a well-written engine could do, this is just blindly postponing calculations for a certain number of frames.

 

[Jeff Graw]

Sadly, that's not limited to Paradox. Many games are still incapable of using more than one, as are many engines. For example, IIRC, Unity only added multicore support in 2018. Hell, even games that do support multiple cores sometimes fuck up and put way more stress on one core than on the other ones. Always take this into account when buying a new machine.

Yes and no. Mostly no.

The Unity API itself is single-threaded. You can't execute a method that references it off of the main thread. But if a method doesn't reference the Unity API, you can do whatever you want with in terms of threading.

What changed recently was the addition of ECS/DOTS, which not only provides a way to use an expanding subset of the Unity API across multiple threads, but also brings with it a high performance subset of the C# language called HPC# + a custom compiler that rivals the speed of extremely optimised C++.

The issue is that this is generally thought of as still not solid enough for production (depending somewhat on what you're doing with it), and certainly works better for something like a massive amount of simple flying cars than, say, AIs for a 4X game.

So something I recently did with Dominus Galaxia, at least for the strategic portion, was that I removed all of the calls to the Unity API from gameplay objects, and where needed created separate graphical objects which do reference the API. So, for example, a fleet is instantiated immediately, while the graphical object for that fleet is instantiated in the next frame update. This let me offload AI processing to a background thread (previously I would need to yield every once in awhile to render the next frame, leading to judder and also slowing down turn processing). AIs still execute sequentially for now, but the goal is to eventually have all the AIs executing at the same time as the player, in the background.

 

[The Brazilian Slaughter]

Unless they ditch shitwitz engine there is nothing they can do. What were they thinking making this game on their crappy engine is beyond me.

A good question, never liked Clausewitz myself. Europa Engine FTW!

Honestly, I think that when you have an engine for immensely complex strategy games, it helps to make it so it doesn't slow down to a crawl as the game goes on. Which Paradox seems to often forget.

That said, CKII does prove Paradox can make a game that doesn't run like crap... eventually. Then again, CKII is old.

Next dev diary is going to be about Performance.

I would say that this is going to be the Moment of Truth for this DLC and this game overall. Almost the entire fanbase is tired of being strung along by Paradox, with the exception of the fanboys.

I think they need a large performance upgrade to get back in the fans' grace. Say, at least +20% performance.

Even 20% is barely enough, I think this game would need to straight-out double its performance in order to actually become fully-functional.

Next dev diary is either going to make or break it.

Let's me honest. It's gonna be break it. They will announce some superficial optimizations that will show 15% "improvement" according to their internal testing. Completely irrelevant for player's experience if it takes several minutes to complete a single in-game year. In the end the game will work even slower because the newest patch/DLC will introduce features that will be even more taxing to the CPU such as Diplomatic Weight, which was already announced to be calculated by counting of all the pops and applying a shitton of modifiers.

The problem with Stellaris is that bad performance is caused not only by bad high-level programming, but also (and probably most importantly) by the aged engine. It's 2020 and the game is still unable to handle several cores properly. Just look at your CPU performance tab in Windows Task Manager while Stellaris is running. One core is used 100% most of the time, while other cores are mostly idle. Ironically, buying a newer PC is likely to make the game slower, because modern machines often use more cores rather than faster ones. But that's not all, in one thread about HoI IV the devs admitted that the engine part which governs the user interface is utter shit, badly designed and performance taxing, but they are afraid to touch it. The UI is so deeply embedded in the engine that working on it, let alone the necessary time and costs, could cause catastrophic cascade. And I guess that this applies to all engine-related problems.

I don't know... its wait and see. I don't think the wider fanbase (AKA not the pdox fanboy squad off Pdox plaza) will take any more of Pdox shit. Either Paradox proves that it can actually make the game not run like crap, or people are going to get tired of it.

I suspect a 20-25% performance improvement is the most we can expect with this patch, given the time/resources/will. Unless someone in their code team strikes it golden and finds out how to give it a massive boost, and they are actually cagey about it because they are actually immensely giddy about the new performance and will use it to overhype Federations and the new patch.

But yeah, your scenario is the one most probable. Pdox fanboys will shrug it off. "Its ok bro, the game runs ok in my NASA computer of a rig. Don't you guys have phones cryo-cooled 5 GHZ core quantum processors?"

Yeah its incredibly bizarre how badly-optimized Stellaris is. In theory its not requirements-heavy but in practice you need a supercomputer to run it well at the highest settings and bigger galaxies. That's... not cool.