LVT reduces strobe cross talk but same latency in overlay

[Chief Blur Buster]

Frame time is not affected by LVT. Display scanout time is affected. You can't measure that with software. You need like a 1000FPS camera and record the screen.

Just so you know, RTSS doesn't always accurately measure frametime per se (pure GPU render time) because RTSS actually measures (GPU render time + frame-cap-blocking time). Which becomes flat with VSYNC ON or RTSS Scanline Sync.

[Kuscheln_Hammer92]

So it could be a+b. What is it measuring while I have two fps and frametimes in the overlay? Is it pulling an average and instant in that cycle or is it giving me the time from one cycle vs the time from the other. It's strange the top number fluctuates with no stuttering or tears but it does when the bottom set do. Especially if both a and b affect smoothness and input delay. I'm thrilled that you came to my aid but feel bad you had to kind of repeat yourself. I've really fallen in love with trying to learn all of this. I currently find best results with back edge sync on. I'm curious though, should I expect to use a higher negative number the larger I made the vbi and LVT? I enjoy back edge sync in most scenarios but still trying to confirm if that has anything to do with the placement of the tearline as you stated. If I could give an example of some really good results I made for myself and then compare/contrast them to some more awful bits. I'm curious if it's just better with a game using Vulkan rather than dx11.

Doom 2016 - 179.97hz - back edge sync @-50 - front porch 57, sync width 8, back porch 355, blanking 420 (nice), and total 1500. Horizontal ended up at 269.955khz. Also don't know how sync polarity with H = positive and v= negative. I just left those the same as that was how it was for my native "240"hz mode the xg2431 came with.

Another success was Cyberpunk 2077 89.717hz back edge sync, front porch 57, sync width 8, back porch 1955, blanking 2020, total 3100

But what has been causing me so many issues due to performance tanking using the scanline sync or RTSS tech besides the overlay is halo MCC. I may have finally resolved it or at least improved it because it's the only one I'm using the enable passive waiting with. I test the polling rate of my mouse ranging from 1000mhz to 4000mhz but that turned out to be irrelevant because that's really only an issue of cpu load? And they all performed the same. I don't know why but it just has awful stuttering and frame drops.
Edit: I had things okayish in halo 4, but halo 2 anniversary still decides to half my frames unless I revert to the old graphics.

I'm sorry about the wall of text. Thank you once more. If it means anything, you've def sparked the passion in another person.

[Kuscheln_Hammer92]

I'm sorry. One more edit: the cyberpunk results were with -140 scanline sync. Sorry about that typo too.

[Kuscheln_Hammer92]

I have really good news. Actually having the sync info up in the overlay has really helped. Didn't realize that's what I was missing. I've been able to make some adjustments that may not be perfect but I'm really happy with so far. Need to look up what the input to present means and if present means as the frame begins or once finished. But with a qvt, that probably doesn't matter as much. I think I read you saying that the placement of the sync line or the tear does effect the qvt? Or it showing up at all? I've been trying to follow your guide making sure it's at the start of the vbi or close enough to it it doesn't bleed into the visible area. It's just been crazy how smooth everything feels. I might use the human benchmark tool just so I can get some kind of idea on input delay but I just have so much joy over the smooth ess of everything right now. I still need to test Halo MCC so wish me luck there. Maybe I'll just throw dxvk 2.1 on it and see what happens.