No. Games living of fancy graphics and physics have very different bottlenecks in the hardware then strategy games like stellaris. The engine has nothing to do with it. Stellaris puts a strain on the CPU, not the GPU
-
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.
While I would say his premise was wrong, the game engine is a major bottleneck for stellaris, it is a decade old and PDX doesn't have anyone who knows how to work on engines in their staff, so there is very little they can actually do to improve it.
PDX needs a new engine, badly.
I fired up the game again after about a year of waiting for performance to be improved post Megacorp (among other needed fixes), only to find that performance is instead noticeably worse. Previously the game ran smoothly at the beginning and then slowed down mid to end game, now I get lag and stutter when panning the camera on max speed from day one. Same computer, same settings, no mods, no extreme size settings. To see no improvement after almost a year and two additional DLC's is very disappointing to say the least.
Last edited:
In early versions my computer ran up to 3000 stars. Now it can barely hold 1000. In the year 2450 it even freezes, requiring a restart of the computer.
According to this thread : https://forum.paradoxplaza.com/foru...ellaris-a-quick-performance-analysis.1138327/
and additionnal tests on my end with the ticks_per_turn command that hint the same, rendering is performed in the same thread as the game logic and most of the threads spend most of their time waiting for tasks to be issued to them.
What's also striking me is that graphically modding the game can divide the tick rate by 2 but my GPU stays at 20% tops and CPU usage stays ~25%, same as vanilla. Consequence I think, of that main thread being bloated to death, solution being separating rendering and game simulation in their own threads, so main simulation thread can take at least a full core if need be and issue work to other threads at a decent speed. So there's at least that for the engine being old.
Then I read several times that assertion about the guys working initially on the engine no longer being a part of the team, thus people at PDox being hesitant to try and alter it, but I could not provide a reliable source for that.
A new engine would take insane amount of work to be viable, and again should be optimized. As above mentioned that even current version could be optimized should some effort be put to it it is obvious that logical solution is to fix what we have rahter than strive for something new and untested.
Maybe this is interesting for someone on here. I think there is plenty to add to this "Megathread unchained" on reddit, just from this thread here alone. I also like the respectfull and nontoxic approach to the matter.
https://www.reddit.com/r/Stellaris/comments/dut80v/the_faults_in_our_stellaris_a_megathread_to/
https://www.reddit.com/r/Stellaris/comments/dut80v/the_faults_in_our_stellaris_a_megathread_to/
Something that has been really counter productive is how toxic some of the posts on this forum can get. I get the frsutration, I really do, but venting it at the devs is definitely not the way forward. The subreddit post is a really postive step forward if you genuinly care about pushing the issues in a way that are more likely to be recieved. I'm really happy to see the reddit admins decided to recognise where the winds were blowing with the core community for this game.
The grand list of unaddressed community issues with Stellaris, very usefull. Thanks.
Although the list is nice to have a summary, even nicer would be a list of Jamor with the addressed community issues 
Most engines are third party these days, which is exactly what PDX needs, as stated, they lack the employees to work on it in the first place, so there is no "working on what they got", and since it is their engine, they can't easily just get someone in to work on it, using a third party engine that has its own team to help would be the most logical move for pdx.
I'm looking forward to news from the developers on their efforts to improve late-game performance and the AI.
Paradox games barely use an engine. Sure, Clausewitz exists, but it's far less substantial than traditional engines like Unity or Unreal. In previous discussions on the topic, devs have repeatedly denied peoples claims of "the Clausewitz engine does X" and "the Cluasewitz engine does Y". In fact, the devs had a rather difficult time explaining what the engine does do, as it's mostly relegated to lower-level stuff.
By and large, the devs appear to mostly be coding things in C++. Discussions of "this is the engine's fault" are almost certainly incorrect.
By and large, the devs appear to mostly be coding things in C++. Discussions of "this is the engine's fault" are almost certainly incorrect.
That's interesting, I didn't know that... Then they do have control over the core parts of the code. This leads me to believe another theory: the devs they have are for new stuff and maintenance, and other things like that. Maybe the people who wrote the core code are no longer with the team, and it's not easy to dive into someone else's code when it's such a complex system like that.
They could expand the team to have enough people to work on modernizing the code (C++ has come a long way since Stellaris was written, and so have compilers, CPUs and GPUs), maybe even breaking the APIs a little if it's for the best (make extensive use of asynchronicity, for example).
There's also the possibility that the people who wrote the code own its source, and they no longer have a contract with Paradox. In that case, yes, I agree with what others were saying when they didn't understand my suggestion (which was that Paradox should buy the rights to a new engine specific for games like Stellaris, not write their own): the amount of work it would take to write the entire system would be just too much.
P.S.: I still think that replacing the text-based thing with a well documented Lua API would maybe boost things up a little, and give a lot more power to mods, reducing compatibility issues that mostly come from the need of replacing Stellaris original files.
Ummm.. I do not think you know how a game engine works.
The problems that are causing these performance issues, are in fact, parts of the engine we do not have access to.
What I understood from his quote is that Paradox's engine is really basic, and Stellaris devs had to implement themselves a lot of stuff game engines provide, thus making Stellaris code bloated and hard to change and optimize (if they had a powerful engine, it would have a team dedicated to keeping it up to date and optimized, while Stellaris devs would just have to adjust a few things when a breaking change would happen and keep working on the game experience).
EDIT: I could have misunderstood it... maybe he actually meant all game engines are general-purpose like Unity and Unreal.
Well he ended with, "Discussions of "this is the engine's fault" are almost certainly incorrect."
So I do not think he had any idea what he was arguing.