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
Seeing as how single-core processors are legacy components that haven't been manufactured in over a decade (even every smartphone created since 2012 has multicore processors), I'm glad the devs now know what multi-threading and multi-core actually is.
Seeing a CPU-bound game, stuck at 8 FPS and 100% maxing out a single core (with the other XX cores sitting idle) should make it readily apparent to anyone why the game must (and now will be) multi-threaded going forward -- especially because the devs intend to add even more features to the game.
100% agree with that. Optimize the single-threaded algorithms to make them as efficient as you can given the time constraints, then throw hardware at the problem (utilize more cores) for the rest. This is exactly what Putnam did for the line-of-sight calculations (optimizations released back in January, then the multi-threading came later / released to prod today).
That's what I was getting at. It's honestly ridiculous how many people demand multithreading instead of basic optimization the single threads first.
More often than not, they'll cherry-pick specific moments in a stream with some player going out of their way to build a massive FPS-sucking monstrosity and act as if the other 99.9% of players will ever bother to reach that.
We all know multithreading was coming, nobody said it wasn't unless they were similarly ignorant to the people acting like it's some sort of "Better Game" button.
Except, you know.... Dwarf Fortress has been multi-threaded for years now. Various different parts of Dwarf Fortress have been handled across different threads for a long time now.
Just some people are ignorant to that and still talk about how it isn't because performance is bad and of course everyone knows if a application is multi-threaded then performance is good..... just ask anyone on Reddit.... *rolls eyes*
For those still catching up on things, the issue has never been that Dwarf Fortress isn't able to leverage multi-threads. It can and has been for various things for quite some time now.
The issue has been that only certain things make sense to be handled in such a way as it stops isolated data in one set of work off-loaded to one thread from being accessible by a workload in another thread.
This means if a crucial calculation needs information being retained elsewhere, at worse if not well coded it can cause crashes and errors, and at best the calculation will pause and sit there waiting for the work being done on the other thread to conclude and for what it needs to be accessible again..... and your 'Improved performance, because Multi-threading' ends up being a reduction in performance.
I like that their response is never something relevant to the conversation and instead just insults people who point out what is actually going on.
Seems to happen a whole lot from some people on here, who apparently don't get flagged for their activity either.
maybe take a closer look at the OP and reevaluate what the tone is
edit: nerd
I assume my FPS gain was so great because my stairs weren't optimal and i prefer large rooms, so the gains i'm getting are definitely in relation to the LOS improvements with multi-threading - loving it so far :D