Re: Driver 441.08: Ultra-Low Latency Now with G-SYNC Support
Posted: 12 Feb 2020, 10:04
Probably their new DCH drivers that was wonky. Didn't autocap FPS in neither CSGO or Apex Legends. Reinstalled with standard drivers instead and it works.
Who you gonna call? The Blur Busters! For Everything Better Than 60Hz™
https://forums.blurbusters.com/
My experience says otherwise. Frostbite3 based games in DX11 mode never seem to respond to LLM's queue manipulation (judging by the obvious effects of the renderaheadlimit dev.console parameter changes but no apparent effects of LLM) but the fps auto-cap of Ultra (usually) works correctly.
Exactly. And yes, now with official G-Sync Compatible certification (fresh official firmware) but I noticed zero change with that last firmware update (other than G-Sync being automatically activated after a clean driver install rather than having to be activated/forced manually for older firmwares --- very old firmwares did have black blinking issues with G-Sync + HDR though but that was with beta drivers and stale firmware during autumn). There is still some mild brightness fluctuation during framerate spikes though (generally only noticeable on loading screens), even in SDR.
I can't remember if I used RTSS or the in-game limiter (but I changed it on the fly, so it wasn't the driver limiter). I think it was the in-game limiter. But that game is a bit jittery in general. I shall try this again with a different game (based on a different engine).
There you go then. I admittedly don't play enough with NULL's auto capping behavior to speak on this with 100% certainty, thus the "Could be wrong on this specific point though" in my last post.janos666 wrote: ↑12 Feb 2020, 11:01My experience says otherwise. Frostbite3 based games in DX11 mode never seem to respond to LLM's queue manipulation (judging by the obvious effects of the renderaheadlimit dev.console parameter changes but no apparent effects of LLM) but the fps auto-cap of Ultra (usually) works correctly.
Probably that specific game/in-game limiter then. And again, it doesn't mean the tearing you're seeing with G-SYNC + V-SYNC "Off" is causing V-SYNC input lag with G-SYNC + V-SYNC "On," just that the V-SYNC function is having to compensate alignment of more frames than it would with, say, an RTSS limit, and/or in another game with more stable frametime performance.janos666 wrote: ↑12 Feb 2020, 11:01I can't remember if I used RTSS or the in-game limiter (but I changed it on the fly, so it wasn't the driver limiter). I think it was the in-game limiter. But that game is a bit jittery in general. I shall try this again with a different game (based on a different engine).
Isn't that typical of G-Sync without V-Sync?jorimt wrote:Even with an RTSS limit? If so, that does seem atypical, unless the frametime was micro-spiking near constantly in that instance, which I find unlikely from your description.janos666 wrote: ↑12 Feb 2020, 09:03What I am wondering about if it's normal to have tearing with a low fps cap and G-Sync without V-Sync. When I testes this, I had to go down from 120 (for 120 Hz) to 60-80 and some tearing still remained (always very close to the bottom of the screen, in the HUD area).
During frametime spikes (from asset loads, etc), yes, typical. But tearing being seen with framerates well within the G-SYNC range (say, 60-80 FPS @120Hz) when there aren't frametime spikes occurring, no, not typical.
Yeah, while the basic recommendations remain the same across VRR implementations (VRR + V-SYNC "On" + minimum -3 FPS limit), it is possible certain non-G-SYNC module software variants of VRR are "looser" in their control of frametime management from frame to frame. This just means said variants probably need the V-SYNC option more (and more often) to prevent tearing within the VRR range than a G-SYNC module does.janos666 wrote: ↑13 Feb 2020, 18:45I need an RTSS limit of 108 or an in-game limit of 109 which RTSS then measures as 107. With these numbers there is may be one tiny hiccup really close to the bottom in every few 10-20 seconds. An in-game limit of 110 which measured as 108 was close to the RTSS 108 result but not as great (those small hiccups were not so rare but relatively still far and few).
Lately, mostly SP games when I have the time; Black Mesa and Hollow Knight atm.spacerocktraveler wrote: ↑19 Mar 2020, 20:06What games do you play and what settings do you have your monitor at?