G sync vs repeating frames.

Talk about NVIDIA G-SYNC, a variable refresh rate (VRR) technology. G-SYNC eliminates stutters, tearing, and reduces input lag. List of G-SYNC Monitors.
SAMURAIWARRIOR13
Posts: 22
Joined: 09 Mar 2018, 14:17

G sync vs repeating frames.

Post by SAMURAIWARRIOR13 » 25 Mar 2018, 00:11

I'm thinking about buying a g sync monitor and have a few questions:
1)
Since g sync matches frames between the monitor and gpu doesn't that mean each frame is only repeat once?
If so, wouldn't playing 30fps on 120hz with each frame being repeated 4 times look much better then 30fps at 30hz with each frame only being shown once?

2)
If I get a monitor that is 144hz and set it to 120hz( via windows display) will I still have to set the cap to 3 fps lower then that so 117fps?

3)
This is probably a dumb question but here it goes,
Does g sync use a backlight similar to PWM technology?

Sparky
Posts: 682
Joined: 15 Jan 2014, 02:29

Re: G sync vs repeating frames.

Post by Sparky » 25 Mar 2018, 02:23

1: if you have something that's exactly 30fps, like a video, then it would look identical. If you're using the 120hz monitor in a low persistence mode(motion blur reduction) then you'll get stroboscopic artifacts due to the mismatch between framerate and refresh rate. If there are framerate fluctuations, the g-sync monitor will avoid stutters.

2: Yes, but if you're using g-sync and want 120hz, you can just leave the max refresh rate at 144 and set your framerate limiter to 120fps.

3) Yes, but that's maybe not the right question. If you're looking for a flicker free backlight they exist, but basically all monitors are going to use PWM technology for dimming. Whether it results in stoboscopic artifacts depends on the PWM frequency and on if there's a low pass filter after the PWM stage of the backlight driver.

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

Re: G sync vs repeating frames.

Post by RealNC » 25 Mar 2018, 07:45

SAMURAIWARRIOR13 wrote:I'm thinking about buying a g sync monitor and have a few questions:
1)
Since g sync matches frames between the monitor and gpu doesn't that mean each frame is only repeat once?
If so, wouldn't playing 30fps on 120hz with each frame being repeated 4 times look much better then 30fps at 30hz with each frame only being shown once?
It looks better on g-sync, because of micro-stutter. 30FPS means there's a frame limiter involved. And that means the frame limiter will be slightly off from the monitor's refresh, and that means you get hiccups every so often.

To do this properly on 120Hz non-gsync, you need to use the Inspector tool and set "1/4" for vsync. This will eliminate the issue, as the game will now lock to a 30Hz vsync signal (it only gets 1 vsync every 4, so 120 becomes 30.) However, even though this completely eliminates stutters, it adds an insane amount of input lag. With g-sync, there will be no added input lag at all.
2)
If I get a monitor that is 144hz and set it to 120hz( via windows display) will I still have to set the cap to 3 fps lower then that so 117fps?
Yes, but that's the wrong setup. What you do with a 144Hz g-sync monitor, is set it to 120Hz for the Windows desktop (so that video playback is smooth), but in the 3D settings in the nvidia panel, you set "preferred refresh rate" to "highest available". This will means that Windows desktop is 120Hz, but when you start a game, it switches to 144Hz. You then set your frame limiter to whatever you want, with the highest still recommended value being 141FPS.
3)
This is probably a dumb question but here it goes,
Does g sync use a backlight similar to PWM technology?
AFAIK, all g-sync monitors (and pretty much all modern high refresh monitors) do not use a PWM backlight. They're flicker-free using a voltage-regulated backlight. However, it still important to double-check before buying a monitor whether it uses PWM or not.
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.

User avatar
Chief Blur Buster
Site Admin
Posts: 11647
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: G sync vs repeating frames.

Post by Chief Blur Buster » 25 Mar 2018, 08:42

SAMURAIWARRIOR13 wrote:1) Since g sync matches frames between the monitor and gpu doesn't that mean each frame is only repeat once? If so, wouldn't playing 30fps on 120hz with each frame being repeated 4 times look much better then 30fps at 30hz with each frame only being shown once?
That is not the purpose of GSYNC.

Repeating frames on a sample-and-hold display does nothing -- 30fps@120Hz looks exactly the same as 30fps@60Hz looks exactly the same as 30fps@30Hz -- GSYNC or no GSYNC -- if the 30fps is already well-framepaced (no erratic frametimes).

The magic of GSYNC happens is fixing random/erratic stuttering. Erratic framerates are smoothed out.

Here's demonstration animations (simulated):
It works best in a stutter-free web browser.
Make sure browser says green "READY". (FireFox or Chrome preferred)
The animation is only accurate when the web browser does not add its own additional stutters above-and-beyond what's in the simulation.

Also GSYNC has no need for "exact frametimes". For GSYNC it's okay for 60fps to sometimes have 1/59sec between two frames, and 1/62sec between two other frames, and it still looks perfectly smooth. You can have random amount of times between frames, and it still can look smooth on a GSYNC or FreeSync display -- there is no need for a fixed interval between refresh cycles on a variable refresh rate display.
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

Image
Forum Rules wrote:  1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
  2. Please report rule violations If you see a post that violates forum rules, then report the post.
  3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!

SAMURAIWARRIOR13
Posts: 22
Joined: 09 Mar 2018, 14:17

Re: G sync vs repeating frames.

Post by SAMURAIWARRIOR13 » 25 Mar 2018, 12:17

I heard that playing 24fps videos at 120hz or 144hz should play smoothly but,
I thought that 24fps videos aren't really 24fps but rather a decimal so how would that divide evenly?
Wouldn't that be a stutter every now and then?
Also
Repeating frames on a sample-and-hold display does nothing -- 30fps@120Hz looks exactly the same as 30fps@60Hz looks exactly the same as 30fps@30Hz -- GSYNC or no GSYNC -- if the 30fps is already well-framepaced (no erratic frametimes).
If that is true then why can't 60hz panels be set to 48hz to play 24fps videos?

User avatar
Chief Blur Buster
Site Admin
Posts: 11647
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: G sync vs repeating frames.

Post by Chief Blur Buster » 25 Mar 2018, 12:21

SAMURAIWARRIOR13 wrote:I heard that playing 24fps videos at 120hz or 144hz should play smoothly but,
I thought that 24fps videos aren't really 24fps but rather a decimal so how would that divide evenly?
Wouldn't that be a stutter every now and then?
On a fixed-Hz display, yes, 23.976 frames per second will occasionally stutter. But for good frame pacing, the stutter is smaller at 144Hz (1/144th stutter vibration amplitude) than at 72Hz (1/72th stutter vibration amplitude)

That said:

There's no such problem with this on a VRR display if the software is properly designed to handle VRR properly. Playing 58.09136 frames per second on GSYNC simply means the display is running perfectly 58.09136 Hertz. The refresh rate can change every frame, every refresh cycle. So random frametime = random refresh time = in perfect sync. The display syncs to whatever frame rate you're at.
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

Image
Forum Rules wrote:  1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
  2. Please report rule violations If you see a post that violates forum rules, then report the post.
  3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!

SAMURAIWARRIOR13
Posts: 22
Joined: 09 Mar 2018, 14:17

Re: G sync vs repeating frames.

Post by SAMURAIWARRIOR13 » 25 Mar 2018, 12:30

I don't think g sync would work for videos because apps like web browsers disable g sync by default and if enabled in web browsers they freak out and cause problems don't they?

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

Re: G sync vs repeating frames.

Post by RealNC » 25 Mar 2018, 12:34

SAMURAIWARRIOR13 wrote:I don't think g sync would work for videos because apps like web browsers disable g sync by default and if enabled in web browsers they freak out and cause problems don't they?
For good video playback you wouldn't a browser. You use a dedicated video player. Some of them support full screen in a way that works with gsync. MPV configured to use OpenGL for rendering is one such player.
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.

SAMURAIWARRIOR13
Posts: 22
Joined: 09 Mar 2018, 14:17

Re: G sync vs repeating frames.

Post by SAMURAIWARRIOR13 » 25 Mar 2018, 12:36

I mean to play stuff like netflix, hulu and amazon video.

User avatar
Chief Blur Buster
Site Admin
Posts: 11647
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: G sync vs repeating frames.

Post by Chief Blur Buster » 25 Mar 2018, 12:51

SAMURAIWARRIOR13 wrote:I mean to play stuff like netflix, hulu and amazon video.
Adding VRR compatibility to such players is extremely difficult, but I think I may have influenced a partial VRR-powered video dejittering breakthrough in this department for non-HDCP-required Netflix players.

Basically, play Netflix with 3:2 pulldown at 60Hz, go through VRR-powered-dejittering via refresh-cycle-offsetting, and then Netflix comes out as perfect 24p.

Keep tuned (ETA: 3Q 2018).
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

Image
Forum Rules wrote:  1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
  2. Please report rule violations If you see a post that violates forum rules, then report the post.
  3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!

Post Reply