Effect of 120Hz on Input Lag

Everything about latency. Tips, testing methods, mouse lag, display lag, game engine lag, network lag, whole input lag chain, VSYNC OFF vs VSYNC ON, and more! Input Lag Articles on Blur Busters.
Post Reply
Stiggosaurus
Posts: 2
Joined: 25 Oct 2014, 20:17

Effect of 120Hz on Input Lag

Post by Stiggosaurus » 25 Oct 2014, 20:39

First off, let me state that I am most definitely not a super-competitive gamer of any sort. I am into sim racing though, and would like to keep input lag as low as possible. But I'm not nearly as sensitive to it as some, and the trade-off of a larger screen is worth it to me. I'm looking into exchanging the initial TV I bought (a 60Hz set) for a 49" Vizio HDTV (M492i-B2) that I suspect can be overclocked to 120Hz (it has a native 120Hz panel). My only reservation is that it's input lag numbers are a decent bit higher (41ms compared to 27ms on my current HDTV)

Assuming that it can indeed be overclocked to 120Hz, how would that affect input lag? This 120Hz TV I'm looking at has measured input lag results of about 41ms. That's right on the border of what I'm willing to accept in a TV. Does running at 120Hz have any impact on input lag? If so, does it make it worse? Better? I'm assuming all the input lag measurements out there are done at 60Hz, so I'm a bit unclear on how a faster refresh rate affects input lag.

Any input would be appreciated!

Trip
Posts: 157
Joined: 23 Apr 2014, 15:44

Re: Effect of 120Hz on Input Lag

Post by Trip » 26 Oct 2014, 06:56

As long as all the other factors stay constant when upping the refresh rate. You will be looking at a drop of around 8ms of input latency. Why?
Well at 60hz the update time of a frame will be 1000/60 = 16.66 ms. Any extra latency is due to additional processing of the image and pixel response time or in case of a strobed backlight the time the display backlight is off. So in this case 41ms of lag means there is 41-16.66 = 24.33 ms of additional processing time for the frame. If this value stays constant when upping the refresh rate (no dropped frames or other funky stuff) then you should be looking at around 24.33 + 8.33 = 32.66 ms of input lag. Though do understand that if you drop in frame rate this wont work anymore so 60 fps at a 120hz screen will feel more laggy on this screen then on the other one. So at best you get around a 6 ms increase of input lag (when you get 120fps). Are you going to notice this? I really don't know that depends on the person playing but do know that precise timings will be affected unconsciously so you will always have to get used to it again.

Stiggosaurus
Posts: 2
Joined: 25 Oct 2014, 20:17

Re: Effect of 120Hz on Input Lag

Post by Stiggosaurus » 26 Oct 2014, 16:43

Trip wrote:As long as all the other factors stay constant when upping the refresh rate. You will be looking at a drop of around 8ms of input latency. Why?
Well at 60hz the update time of a frame will be 1000/60 = 16.66 ms. Any extra latency is due to additional processing of the image and pixel response time or in case of a strobed backlight the time the display backlight is off. So in this case 41ms of lag means there is 41-16.66 = 24.33 ms of additional processing time for the frame. If this value stays constant when upping the refresh rate (no dropped frames or other funky stuff) then you should be looking at around 24.33 + 8.33 = 32.66 ms of input lag. Though do understand that if you drop in frame rate this wont work anymore so 60 fps at a 120hz screen will feel more laggy on this screen then on the other one. So at best you get around a 6 ms increase of input lag (when you get 120fps). Are you going to notice this? I really don't know that depends on the person playing but do know that precise timings will be affected unconsciously so you will always have to get used to it again.
This is exactly what I was looking for, and makes perfect sense. I had comes across the 16.66 and 8.33 ms numbers for 60Hz and 120Hz respectively, and suspected that there may be some drop because of that fact. Your explanations clears it up--thanks!

I should be able to maintain a framerate over 120fps for the most part, but there's no guarantee that the display I'm looking at even can be overclocked to 120Hz. I'm not by any means dissatisfied with what I'm seeing on my TV I just bought, so I think I'm just going to stick with what I've got. Definitely appreciate the explanation!

flood
Posts: 929
Joined: 21 Dec 2013, 01:25

Re: Effect of 120Hz on Input Lag

Post by flood » 26 Oct 2014, 18:38

Trip wrote: Well at 60hz the update time of a frame will be 1000/60 = 16.66 ms. Any extra latency is due to additional processing of the image and pixel response time or in case of a strobed backlight the time the display backlight is off. So in this case 41ms of lag means there is 41-16.66 = 24.33 ms of additional processing time for the frame
that depends on what exactly is measured when someone reported 41ms for that monitor.
in general it's hard to say how much lower input lag would be by 60hz -> 120hz but probably it's safe to say that it would be between 41 ms and half of 41ms.

User avatar
RealNC
Site Admin
Posts: 3741
Joined: 24 Dec 2013, 18:32
Contact:

Re: Effect of 120Hz on Input Lag

Post by RealNC » 27 Oct 2014, 02:28

With vsync off there is no difference in input lag. With vsync on, 120Hz will have half the input lag of 60Hz.
SteamGitHubStack Overflow
The views and opinions expressed in my posts are my own and do not necessarily reflect the official policy or position of Blur Busters.

Trip
Posts: 157
Joined: 23 Apr 2014, 15:44

Re: Effect of 120Hz on Input Lag

Post by Trip » 27 Oct 2014, 05:12

I don't know if my previous comment is correct about the 24 ms staying constant. If the refresh rate is doubled it could be frame processing is speeded up too. Since depending on what is processed it might be a frame will be buffered and will just be processed faster moving on to the next buffer. Resulting in less input lag then what I said before. So what is going to happen?
Nobody knows until you test it. A possible scenario the screen buffers two images for processing since 41%16.33 = 2. The rest of the latency could be due to pixel response time 41-(2*16.33) = 8.33 ms. So it could be that input lag can be reduced to 2 * 8.33 (buffered frames) + 8.33 (pixel response) = 24 ms.
But nothing is sure until you test this to see how the screen responds to this refresh rate. Maybe some expert might have a better explanation about this though.

Post Reply