Blur Busters Forums

Chief Blur Buster wrote:From what I am understanding in newer developments [...]

Since Nvidia hasn't publicly announced any module version past "2.0" a few years back, it's difficult to tell if there have been any further refinements to these aspects in the modules featured in the more recent G-SYNC monitors.

That said, I'm all for them simplifying the LFC and VRR range functionality (low, mid, and high) with the continually developing VRR standards, even if that means software will ultimately replace some or most hardware functionality.

I can see a software/GPU-side solution eventually making module-based LFC unnecessary, and I can even see them tightening up or eliminating any "polling rate" (if it still indeed exists) to improve response time to frametime variances in the upper/lower range, but no matter the method, extreme frametime variances/spikes are still problematic when trying to achieve zero-tear VRR on <240Hz monitors, while at the same time, avoiding the VRR's extension of the existing frametime spikes (when compared to standalone V-SYNC or V-SYNC OFF in the same instances), due to the more limited static scanout speeds when compared to the eventual 480Hz and up, where this issue (and input lag in general, VRR or no VRR) will obviously become less and less impactful as static scanout speeds (and hopefully average framerates) increase.

Of all the requirements -- dynamic overdrive is the biggest difficulty.

It's horrendously complicated, especially in varying frametimes.

The overdrive needs to try to predict the next frametime, not the previous frametime, which is sometimes hard to guess. If you can't do dynamic overdrive properly -- an display engineer can easily get artifacts worse than fixed overdrive. To most people, it feels like voodoo science -- the art of overdrive tuning. Good dynamic overdrive required a powerful FPGA chip initially...

^ Yeah, that definitely as well.

I was going to mentioned VRR overdrive implementation in a version of my previous post, and how difficult it would be to get right if they went from module to GPU or software side-only, but I ultimately decided to stay in my lane, and cut it

Was there mentioned that game engine can easily ignore that driver option ("pre-rendered frames") value?
Is there a way to check it?

Chief Blur Buster wrote:

pwn wrote:But based on the conversation, I understand the best configuration for playing CsGo is Locked FPS 237 (I have a 240Hz monitor supporting Gsink) and not free FPS 300+?

Yes, but only if using G-SYNC (what you called Gsink)

If you use VSYNC OFF instead, use a higher cap (e.g. 300fps or 500fps), to gain Advantages of framerates above refresh rates.

You do have a choice of:
1. G-SYNC + 237fps cap
2. VSYNC OFF + higher cap (300fps or 500fps)

Thanks for the answer.
I understand correctly that the most optimal way to restrict FPS is RivaTuner?

Q83Ia7ta wrote:Was there mentioned that game engine can easily ignore that driver option ("pre-rendered frames") value?
Is there a way to check it?

You can check it easily. Set it to 8 in nvidia inspector and then play with vsync ON at 60Hz. If you don't get massive input lag, then the game somehow is ignoring the setting.

RealNC wrote:You can check it easily. Set it to 8 in nvidia inspector and then play with vsync ON at 60Hz. If you don't get massive input lag, then the game somehow is ignoring the setting.

Are sure that 8 is value that driver will accept? May be max is 4 and if value > 4 then driver will set is to auto (3 iirc).

Q83Ia7ta wrote:

RealNC wrote:You can check it easily. Set it to 8 in nvidia inspector and then play with vsync ON at 60Hz. If you don't get massive input lag, then the game somehow is ignoring the setting.

Are sure that 8 is value that driver will accept? May be max is 4 and if value > 4 then driver will set is to auto (3 iirc).

8 works fine here. But if you're paranoid, you can use 4 and then 1. You should still be able to tell the difference.

But 8 should work. I used it in the past for tests.

RealNC wrote:8 works fine here. But if you're paranoid, you can use 4 and then 1. You should still be able to tell the difference.
But 8 should work. I used it in the past for tests.

Thanks! Will test soon!

Tested on Quake Champions. 60hz, vsync ON (nvidia cpanel and game). Maximum pre-rendered frames 1 or 8 gives same result. So game ignores it. While default value is auto (equals 3) it doesn't give disadvantage for new players and it's good for any multiplayer competitive game.