These games absolutely exist but I have to say after a couple years of g-sync gaming it's been a pretty rare issue. A lot of games require 60fps though, like Bethesda's games, but you can just cap them at 60 with something like RivaTuner and keep v-sync off.

yeah thx
ive found tripple buffer in radeon app but it is only for opengl as i see and most games seems to be using directx and a few uses vulkan

I've done some tests on Quake 2 RTX. I think it's slightly better with triple buffer enabled, but not as good as without V-sync. So yeah, it helps a bit, the input lag is a little smaller, but still not perfect.

For them i'm not sure. IMO, it should be mandatory for all cards to have basic support. It's more or less double buffering for the cards themselves.
What are you getting with it and what are you getting without it? How much lag?

If it is an option I will always turn v-sync on. Torn frames are distracting and while I prefer 60 fps it's not going to kill me if the game runs lower, to a point. I played Witcher 3 at 40 fps (the only fps cap that worked, which I had to force over the NVIDIA control panel) and never really noticed I was not playing closer to 60 whereas when I was getting higher but uncapped frames the camera control was jittery and not something I could put up with.

It's not always an option, or at least there are some games where no matter what I try I cannot force it (Drakensang physical), but if I can I will always prefer v-sync for cleaner image and smoother control.

Back when I was growing my PS3/360 collection more the choice for a multiplat title was sometimes down to v-sync (Sony at some point seemed to make it a policy whereas Microsoft did no). Some games ran at higher frames on 360 but the PS3 had a cleaner image (except when talking about resolution, that was generally also a 360 thing). Binary Domain and I think Dark Souls 2 were games like that. It helped some exclusive titles hold up well on the PS3 as well, such as Resistance 3.

How can I measure lag? :) These are my feelings only. The input lag is slightly lower with triple buffer enabled, but it's still there.

There are tests. Like the UFO Test.

True, but then you need a GPU and a display device that both support it. G-Sync vs FreeSync etc.

As someone who doesn't currently have access to either technology, I guess I should check what RivaTuner does for me. However, I got the impression from some pages that unless you are getting constantly over 60 fps with your game(s), you should not use RivaTuner, like that it might be even worse than playing e.g. a 55 FPS game with regular vsync (which would lock it to constant 30 fps I guess)? Or did I misunderstand that?

To me it is both. If I play something like Dungeon Siege 1 or 2 which would run up to 200 fps without vsync, I'd like to limit to constant 60 fps with either vsync or RivaTuner, whichever does it best.

But if I play something like The Witcher 3 or Horizon Zero Dawn on the same laptop, the best I can hope for is maybe semi-constant 30 fps (with vsync on), so I am unsure if RivaTuner offers anything of use there, compared to running those games either with vsync off, or vsync on.

This is all still so confusing... I really loved that one RivaTuner configuration page which suggested how you first calculate the _exact_ refresh rate of your display device with like 5 decimals, then round that down and/or substract 0.001 from that number and enter that to RivaTuner... and then do more optimizations based on what your results are. Wow. Anything for smooth and lag-free gaming...

Yes. See my previous post attached picture. You can limit the frame rate to any numer, including decimals or use the Scanline Sync.
Note that it limits the framerate on any aplication (Directx 9+ and OpenGl, don't know about others), so you may want to have profiles for individual games.
You don't need precise calculations for limit the frame rate, although, to be virtually tearing free, those precise numbers seem to work best.
This link makes it easy to check the actual frame rate, if the browser V-Sync implementation is good.

Battery + wires + split bobby pin (actually broken into 2 pieces) + stick + small resistor + LED = pressure switch. The idea is that you want to make it so that when you press the key down, it makes contact with the split bobby pin which are connected to wires that are connected to LED, resistor, and battery, making a pressure switch. When you press the 2 ends together they'll light the LED, but since you'll have a stick between the two they'll immediately separate again when they try to bend back into place. For all intents and purposes, this would activate the LED so fast that i don't believe a slow motion camera exists that would be fast enough to capture the delay on the light.

Place pressure switch under keyboard key or mouse, record screen with camera phone at the highest FPS, use ffmpeg to convert video to individual frames. People like to do this with shooting, but movement keys on the keyboard work better. Repeat the test at least 10 times and average the results.

The problem with this is that it can desync and there'll be some slight tearing as it does. Most reliable (and efficient) solution would be using a driver that actually found a way to get a vsync IRQ (i'm sure the manufactuers have done it).

It's not a complex topic, really. The basic idea is that the screen refreshes every so often, then goes into a wait state between refreshes. Slow Mo Guys (or whatever they call themselves) has a good video captured with a slow motion camera showing how this actually works. The trick is that you have to send all the data to the frame to the video card (and then the screen) while it's *NOT* doing the vrefresh, else you end up with a tear. Now, the part that gets harder is that data transfers like that are not instant, and a v-refresh can occur while the data's being sent to the TV, so ideally, your GPU wants to send the data immediately after the refresh is done, or right after it's started (if the GPU is slower to send than the TV is to receive, causing only 1 frame lag (actually slightly less) to give it maximum time to send the data without another refresh occuring while it's sending.

I'm playing Witcher 2 atm and I tried turning off V-sync in the graphics settings. It didn't make the slightest bit of difference - the game is still very smooth, with no tearing. I'll turn it off from now on by default, unless I'm having issues.

That's way too much of a hassle. Not planning to construct a machine that will measure the input lag. I'm not that determined to research the issue. :)

And you don't feel any input lag? Check some FPS game. The result my vary depending on the game.

i don't, because it reeks havoc on low end GPUs.

And this is why people like me don't believe people like you. People make all sorts of claims giving advice on technology, but they can't even do the bare minimum to verify their claims. They just defer to someone else who's complaining, and it becomes indistinguishible from mass hysteria. Until someone can verify to me that a video card and driver combination that is working properly actually introduces significant input latency, i'm just going to go with people turning around and saying "I couldn't throw a strike 'cause the wind was blowing." Without some sort of physical evidence that something is there, i'm just going to assume it's all in your heads (we've only been working with that standard for the past 200 years). I don't. And i do play them from time to time (i especially like fallout and i play ranged characters in elder scrolls, but i'll even play the twitchy fraggers here and there). I get more input lag from games using a hard disk drive than i do from vsync.

The majority of people posting in this thread claim that the lag is present and that's why they don't use V-sync. If you don't believe it, that's your problem. I'm not planning to build a device that measures lag only to satisfy your doubts. Sorry.

Then explain to me why I'm getting better results in online shooters when I turn V-sync off? Is it only my head again?

Skyrim is another great example. I have a massive mouse cursor lag in this game, but it disappears almost completely when I look around with my character. So only the mouse cursor is affected in a meaningful way.