I can see it yeah. But why is adaptive sync on when you are running a fixed framerate? Or did you do this just to illustrate? I can imagine adaptive sync is averaging the FPS across a period of time and ends up alternating between 114-130, because the numbers are 'around 120' and if you add enough of those variations up, you can still average 120
In other words, frame cap may disrupt it. Its also well possible that your actual FPS is not 120, its just what the game says it should be. Might very well fluctuate a little or deviate from the actual number. (If you want evidence of that, check out Unigine Valley FPS counter alongside the RTSS one. Good fun! The clockspeeds too. They're always off)
Perhaps your monitor can use a firmware update or there is a known error with it? Seems very much like a monitor/settings issue, that, OTOH its not really common to use that combo of settings. Like you say, we tend to force Vsync with a Gsync mode, because that way Gsync can take control of the buffer. It is then vital you don't use anything on top of that to control framerate or refresh rate.
So far I can't rule the monitor out. If you've tried different combinations and nothing is pleasant or stable in monitor refresh rate, something is amiss. But to verify that, you need to repeat your tests in another game. One game cannot tell you anything about the monitor and
@nguyen pointed out its a game specific thing.
One thing I'm seeing with your analysis... you jump from one place to another. Now you're doing framerate caps with an ingame limiter, but that is the worst possible solution for accuracy.
Try to always approach things in layers because that is how PCs work.
- Layer 1 (closest to metal): Nvidia Control Panel, these are basically hardware toggles.
- Layer 2 (closest to OS): RTSS on screen display/Afterburner. Still software though that reads out sensor data and sends GPU commands, so has timing/latency impact (you're looking at what has been).
- Layer 3 (closest to application): in-game settings.
If you force things in a lower layer, it will apply to all higher layers. But, higher layers can still interfere with the pipeline. Example: You set Vsync in NVCP, but then cap the FPS to 25 in RTSS or in the game. The FPS cap will kick in before Vsync can do the work, but Vsync is still clogging up your buffer.
So, try to always work in/from a single layer, the lowest one possible. If you FPS cap, you can now even do it in NVCP, so do it there and don't touch anything in RTSS or the game. If you Vsync in NVCP, you keep all Vsync in game OFF. You are trying to find the culprit here within the millisecond: work with similar precision. Every tiny hiccup will screw with your frametimes.
Also, a general note about watching FPS counters, performance and especially crushing stuttery experiences... your mind will screw with you bigtime and it can become hard to unsee something you've put into it. At that point... take a break, try to set your mind to something else, come back to it and then just start gaming, try to forget it and analyze what you've just done. Was it truly a bad experience? Or is this a mountain out of a mole hill? This is also why you really need to start focusing on OTHER games. Warzone is just a random shooter don't make it bigger than that. Can't expect everything to run perfect.
Thing is the general smoothness of your gameplay evne in that video with all the fluctuation of refresh seems buttery. Its fine and that number really is just a number, perhaps.