I don’t know what VRR windows you get on handheld displays these days, but at 30fps it shouldn’t be super useful compared to vsynced 90Hz. 90 is 11ms intervals for your next frame, and if you’re pushing the hardware at ~30fps you may have bigger swings between frames in VRR (e.g. you could have 8ms between two frames and 28ms between the next two), which still reads as stutter, with or without VRR.
So it’s not as much of a no-brainer as you may think. That’s basically the same reason Lenovo insiders gave for why the 1600p 144Hz panel in the Lenovo Legion GO is also not VRR. In that case it makes a bit more sense because that’s just 7ms between refreshes, so you may genuinely struggle telling the difference between that and VRR if you’re rendering less than 60fps.
I think Digital Foundry does a lot of good advocacy and educational content, but sometimes they get hung up on pet peeves and give people the wrong impression about which buzzwords are important on which contexts.
Of course Lenovo are going to have marketing for why them saving money is actually better for the consumer. That is just how marketing works. If Valve were at all competent at it, they would be doing the same.
VRR doesn’t stop stutter. But it helps a lot when you have those gradual fluctuations. Think “If I look up, my FPS drops by 20%”. At which point you no loner have to worry at all about multiples to avoid screen tearing or all of that annoying stuff. All of which is REALLY nice when you are at the limits of your system. Whether that is pushing 100-144 FPS or 20-40 FPS. It won’t make it look like it is running perfectly, but it very much helps a lot and there is a reason that VRR is one of those “most noticeable hardware improvements” you can get.
And can we please skip out on the “Oh, people just don’t notice improved graphics and refresh rates anyway” nonsense? I realize the Steam Deck is a handheld, but this isn’t a Nintendo Switch thread.
FWIW, LTT seems to have asked about VRR and they hypothesize, based on the answer, that they’re sourcing from the same place as Nintendo and that is limiting the VRR option: https://www.youtube.com/watch?v=uCVXqoVi6RE
But my point stands in that you’re thinking about the target spec of the display, not the games. There IS a difference between 20-40 and 100-144 fps. First, because it’s a lot harder to keep a steady rate at 7ms frame budgets and second because the sense of stability doesn’t have the same demands.
And yes, it’s a perceptual thing. Some people will be more sensitive than others, but I would feel comfortable showing a 28-30 fps clip to people on a 144Hz vsync and a VRR display and asking them to spot which is which. Simply put the gaps in miliseconds between those two things are going to be too similar to tell apart. I know because I’ve tried. I have 100, 120, 144 and 165 fps displays, both VRR and vsynced. I’ve messed around with this for a long time for fun and profit.
I have no question that VRR would be a slight improvement, but I’m also not surprised that at these levels of speed and size both Lenovo and Valve decided that it wasn’t worth to chase VRR compared to the high refresh alternative. That gels with my own experience.
I don’t know what VRR windows you get on handheld displays these days, but at 30fps it shouldn’t be super useful compared to vsynced 90Hz. 90 is 11ms intervals for your next frame, and if you’re pushing the hardware at ~30fps you may have bigger swings between frames in VRR (e.g. you could have 8ms between two frames and 28ms between the next two), which still reads as stutter, with or without VRR.
So it’s not as much of a no-brainer as you may think. That’s basically the same reason Lenovo insiders gave for why the 1600p 144Hz panel in the Lenovo Legion GO is also not VRR. In that case it makes a bit more sense because that’s just 7ms between refreshes, so you may genuinely struggle telling the difference between that and VRR if you’re rendering less than 60fps.
I think Digital Foundry does a lot of good advocacy and educational content, but sometimes they get hung up on pet peeves and give people the wrong impression about which buzzwords are important on which contexts.
Of course Lenovo are going to have marketing for why them saving money is actually better for the consumer. That is just how marketing works. If Valve were at all competent at it, they would be doing the same.
VRR doesn’t stop stutter. But it helps a lot when you have those gradual fluctuations. Think “If I look up, my FPS drops by 20%”. At which point you no loner have to worry at all about multiples to avoid screen tearing or all of that annoying stuff. All of which is REALLY nice when you are at the limits of your system. Whether that is pushing 100-144 FPS or 20-40 FPS. It won’t make it look like it is running perfectly, but it very much helps a lot and there is a reason that VRR is one of those “most noticeable hardware improvements” you can get.
And can we please skip out on the “Oh, people just don’t notice improved graphics and refresh rates anyway” nonsense? I realize the Steam Deck is a handheld, but this isn’t a Nintendo Switch thread.
FWIW, LTT seems to have asked about VRR and they hypothesize, based on the answer, that they’re sourcing from the same place as Nintendo and that is limiting the VRR option: https://www.youtube.com/watch?v=uCVXqoVi6RE
But my point stands in that you’re thinking about the target spec of the display, not the games. There IS a difference between 20-40 and 100-144 fps. First, because it’s a lot harder to keep a steady rate at 7ms frame budgets and second because the sense of stability doesn’t have the same demands.
And yes, it’s a perceptual thing. Some people will be more sensitive than others, but I would feel comfortable showing a 28-30 fps clip to people on a 144Hz vsync and a VRR display and asking them to spot which is which. Simply put the gaps in miliseconds between those two things are going to be too similar to tell apart. I know because I’ve tried. I have 100, 120, 144 and 165 fps displays, both VRR and vsynced. I’ve messed around with this for a long time for fun and profit.
I have no question that VRR would be a slight improvement, but I’m also not surprised that at these levels of speed and size both Lenovo and Valve decided that it wasn’t worth to chase VRR compared to the high refresh alternative. That gels with my own experience.
Here is an alternative Piped link(s):
https://www.piped.video/watch?v=uCVXqoVi6RE
Piped is a privacy-respecting open-source alternative frontend to YouTube.
I’m open-source; check me out at GitHub.