Asenna Steam
kirjaudu sisään
|
kieli
简体中文 (yksinkertaistettu kiina)
繁體中文 (perinteinen kiina)
日本語 (japani)
한국어 (korea)
ไทย (thai)
български (bulgaria)
Čeština (tšekki)
Dansk (tanska)
Deutsch (saksa)
English (englanti)
Español – España (espanja – Espanja)
Español – Latinoamérica (espanja – Lat. Am.)
Ελληνικά (kreikka)
Français (ranska)
Italiano (italia)
Bahasa Indonesia (indonesia)
Magyar (unkari)
Nederlands (hollanti)
Norsk (norja)
Polski (puola)
Português (portugali – Portugali)
Português – Brasil (portugali – Brasilia)
Română (romania)
Русский (venäjä)
Svenska (ruotsi)
Türkçe (turkki)
Tiếng Việt (vietnam)
Українська (ukraina)
Ilmoita käännösongelmasta
Been looking forward to this functionality since I have 2 screens (a monitor and a tv, both with different refresh rates). I usually game on the tv.
Testing this version, even when I select the tv as output, LS seems to sync only to whatever is the main display -in this case, the monitor's refresh rate (both Draw FPS option in LS and PresentMon show it doing so) and there's visible judder in the upscaled 3D app because of desync.
Is there a way to make LS "sync" to the screen refresh rate it is upscaling to?
There should be no "desync" as game capture and scaled content rendering happen independently. So it's caused by something else.
As for the refresh rate, I haven't worked out the details myself yet. If I scale the game from a 60 Hz monitor to a 120 Hz monitor, LS runs at 120 FPS. If I keep the game on 120 Hz monitor and output to it, then for some reason it is 60 FPS. The 60 Hz monitor is the primary one. What is your configuration?
If I start a game (configured to start on the TV), the game's window starts on the TV, no problem. When using LS to upscale, it upscales without a problem, but as I mentioned, there's judder and both internal LS framerate counter and PresentMon show LS running at 75 fps.
If I make the TV the main screen, judder disappears and both LS framerate counter and PresentMon show LS running at 60 fps, that's why I mentioned "desync": in my limited knowledge about the topic my theory is LS is synchronizing the output to the main screen refreshrate (based on LS Draw FPS and PresentMon info), not to the screen it is upscaling to.
I've seen the same behaviour with RTSS's hybrid s-sync: if the "wrong" screen is synced to, there's judder.
This contradicts my testing. As I already said, my primary screen has 60 Hz, while LS runs at 120 FPS on the secondary screen.
Have you tried keeping the game on the monitor and scaling on the TV?
Another test: scaling a game from 120 Hz to 60 Hz monitor uncaps LS framerate (I get over 120 FPS) ¯\_(ツ)_/¯
Here's a screenshot[i.imgur.com] showing the Pendulum Demo running at 120 on the monitor (no vsync, just a plain framerate cap at 120) and you can see both PresentMon and LS's Draw FPS option showing 75 FPS on the TV.
I removed both Nvidia CP and RTSS profiles for LS just in case those were affecting the test.
Now, if I launch the Pendulum Demo and then make the TV the main display and then upscale with LS, here's how it looks[i.imgur.com].
It definitely has something to do with the primary display, but I don't see a consistent pattern of behavior yet. Without knowing what Windows does under the hood, it will be difficult to figure it out. By the way, what version do you have?
I also ran both scenarios with the new "Optimization for windowed games" option disabled, but the results are the same.
Hi,
How do you measure the latency?
Hi handsome, with the in app Draw FPS , I use this option since you release it and I love it, but as said, in this last version I notice an increase in latency, no matter the scaling type or the preferred GPU, laptop here, before when I was using intel gpu (integrated) with bicubic CAS or FSR the latency was in between 2 and 3 ms, and high fps value, while using nvidia (dedicated) I was getting not more than 30 fps no matter what and latency around 10-13ms, now with this last version, I get ok ish fps with both, higher fps with integrated as usual, but with both (intel integrated, nvidia dedicated) the latency is 10ms +
App draws the frame time (time it takes to scale one frame), not latency. But it may also affect latency when it's high. Does "double buffering" option makes a difference ?