VT and input lag

Everything about displays and monitors. 120Hz, 144Hz, 240Hz, 4K, 1440p, input lag, display shopping, monitor purchase decisions, compare, versus, debate, and more. Questions? Just ask!
b0t
Posts: 108
Joined: 16 Jan 2017, 00:10

Re: VT and input lag

Post by b0t » 03 Apr 2022, 17:29

Chief Blur Buster wrote:
03 Apr 2022, 16:54
b0t wrote:
01 Apr 2022, 16:30
so if VSYNC OFF + Large VT doesnt benefit me
Only if you're not using strobing or a modified VSYNC ON

While there are monitor-specific exceptions, these are general rules of thumbs:

Things that benefit from QFT:
- Strobing with ANY sync technology
- Non-strobing with a modified VSYNC ON clone (e.g. RTSS Scanline Sync)

Things that do not benefit from QFT:
- Non-strobing with ordinary Microsoft Windows VSYNC ON (due to its beginning-of-VBI presentation behavior).\\
b0t wrote:
01 Apr 2022, 16:30
VSYNC OFF + strobing + QFT/Large VT like in my case, benefits the crosstalk more than the latency reduction, I actually am seeing improvement in the bottom rows of Alien Invasion in Moving Photo@UFO;
Correct.
b0t wrote:
01 Apr 2022, 16:30
but question is, doesnt it actually ADD input lag/latency somehow? considering Im not using any SYNC technology like u previously asked (only strobing+the 1875vt like I already said)
Since VSYNC OFF randomizes the splicing of the signal, average input lag of QFT vs non-QFT is identical. However, latency gradient does become slightly distorted. The input lag of top versus center versus bottom.

The interaction between signal scanout, display scanout, and global strobe flash, can distort the latency gradient (the lag difference of gametime:photontime for top edge of screen versus bottom edge of screen).

Nonstrobed + VSYNC ON creates input lag gradient where top < center < bottom
Nonstrobed + VSYNC OFF creates identical lag gradient where top = center = bottom
Strobed + VSYNC ON creates identical input lag for top = center = bottom
Strobed + VSYNC OFF creates input lag gradient where top > center > bottom

Adding QFT simply tightens the latency gradient, where DIFFERENCE(top, bottom) is simply the time delta between delivery of first pixel row and delivery of last pixel row. However, in the case of VSYNC OFF, the AVERAGE of (top, center, bottom) is unchanged, even if DIFFERENCE(top, bottom) is a smaller lag difference, since top can increase a bit while bottom decrease a bit.

So yes, you are seeing modified latencies, depending on where your eyes are looking at (crosshairs? Top edge? Bottom edge?)

You just simply have to get used to the new latency mechanics -- it is a matter of aimtraining to your new settings.
Thank you for the detailed explanation.

my FPS in-game with 3rd party client+anti cheat appears to be even lower than I thought, so I lowered my refresh rate even further to 120Hz.
I created a new custom res with VT 2250 , lowered the Area value to 0 and ... Im actually speechless.

screen appears to be completely clean in the Moving Photo UFO Test... id say that only the last (bottom) row of ufos is MAYBE 20% worse than rest of the screen, but the proportions are amazing.

too bad I must play on such low and pixelated resolution in-game for the next couple of months untill I get my new pc. (hopefully paired with a new XG431 ;) )

Post Reply