Install Steam
login
|
language
简体中文 (Simplified Chinese)
繁體中文 (Traditional Chinese)
日本語 (Japanese)
한국어 (Korean)
ไทย (Thai)
Български (Bulgarian)
Čeština (Czech)
Dansk (Danish)
Deutsch (German)
Español - España (Spanish - Spain)
Español - Latinoamérica (Spanish - Latin America)
Ελληνικά (Greek)
Français (French)
Italiano (Italian)
Bahasa Indonesia (Indonesian)
Magyar (Hungarian)
Nederlands (Dutch)
Norsk (Norwegian)
Polski (Polish)
Português (Portuguese - Portugal)
Português - Brasil (Portuguese - Brazil)
Română (Romanian)
Русский (Russian)
Suomi (Finnish)
Svenska (Swedish)
Türkçe (Turkish)
Tiếng Việt (Vietnamese)
Українська (Ukrainian)
Report a translation problem
^ this. Or parallel computing for that matter. If you really want to learn something OP, check Amdahl's law for example to get a better understanding why something like you requested is very hard to do and / or almost not possible with mods.
You will need to wait until both the OS and the hardware are all efficiently running parallel.
So I agree with the other posters.
The game would massively benefit from being re-written at the engine level however by some very talented programmers Egosoft could outsource the work to. As the current way the game handled things is VERY poorly optimized, and the game's engine is basically held together with duct tape as it is.
So I never understood what people are trying to say when they say the game only uses one core.
Sure, one thread is going to be the main loop thread, and it's the devs' job to not overload the main thread at whatever target fps (usually it's going to be 60). But that's how all game engines work.
At some fps number, assuming infinite gpu horsepower and infinite memory xfer speed, every game will cpu bottleneck on one of the cpu threads. If you make a toy engine that does basically nothing besides render a quad canvas, you'll cpu bottlneck at something like 10k fps because the rendering thread reaches 100% cpu utilization of one core -- and there's no way to split that thread up. Does that mean that toy engine "only runs on a single core"? No.
For some applications that's useful, so you can still claim that your quad core has four times the performance of a single core (or whatever), but in practice many applications can't use multiple cores at all or the extra work of implementing the logic for distribution of tasks and synchronization of results isn't even worth the minor performance increase.
Think of it like going from a single employee to a team. You can't cut down dev time by 4 just by adding 3 other guys to a project. In fact, you may well end up slowing the project down because not all sub tasks can be split up and handled by different people. And the single guy that worked alone before now has to spend alot of time discussing, reporting and documenting instead if doing his actual work.
But I wouldn't call it a scam, otherwise we'd all still be running Pentium 4's OC'd to 5ghz.
There ARE things that are highly parallelizable -- the main loop thread (usually also gets the user input), rendering thread, audio thread, UI update thread, AI thread, physics thread (if there are any)... that's 6 right there, and then you could easily add a 7th for very low priority background stuff (stuff that doesn't matter if it takes 10 seconds to get done).
I'm talking about real physical cores, not hyperthreading/SMT -- that's highly dependent on how something is coded and not really something worth optimizing for.
I spent alot of time playing this game and don't want to go back because the end game performance issue. Always = 20-30fps unplayable slide show making it feel like you wasted your time.
It is a hardware issue then an OS issue before the Game dev gets any input