Does matching polling rate to monitor Hz make sense or no?

Everything about displays and monitors. 120Hz, 144Hz, 240Hz, 4K, 1440p, input lag, display shopping, monitor purchase decisions, compare, versus, debate, and more. Questions? Just ask!
RocketRager
Posts: 21
Joined: 05 Apr 2021, 15:35

Does matching polling rate to monitor Hz make sense or no?

Post by RocketRager » 06 Apr 2021, 15:25

Game:Rocket League
Controller: Dualsense but also used Dualshock
PC: Ryzen 2600, RTX2070, Monitor AOC 1440p 144Hz, RAM 16gb Gskill 3200, Windows 10

Settings as of today, from memory since I’m at work (but have tried many)1440p, full screen, in game frame cap at 141, gsync on, v sync on, advanced frame rendering 1, low latency mode on. Can’t think of anymore.

FPS stays around 250-300 most of the time.

I have struggled since sept. to find settings that feel good and consistent. When controller is OC, of course it feels better.

The main problem, I am microseconds late to challenge the ball frequently, I have well over 2k hours in the game and “know how to play it right”. But having my replays analyzed by other people they simply tell me I’m always late to challenge, and late to the ball when I do challenge. Of course sometimes I arrive first. My rank is higher than average but not much. And I’m stuck there. Zero progress is being made. Matter of fact a lot of the time I will lose huge chunks of rank in one night, and have to grind weeks maybe to get it back.

It hit me based off a comment chief made in another thread about 1k polling mice not working well with certain Hz monitors or certain FPS

Would it be better for me to DOWN clock the controller to better match my 141 limited frames? I don’t even know if that makes sense to anyone but it seems reasonable to me.

This is a strange world I’ve just entered. I’m seeing tons of posts that say 240-280 hz is like cheating I’m this game. If it’s that simple, I guess I’m buying a new monitor....

Futuretech
Posts: 35
Joined: 11 Oct 2020, 23:52

Re: Does matching polling rate to monitor Hz make sense or no?

Post by Futuretech » 06 Apr 2021, 15:51

This concept you mention has been around since the quake era back in mid-late 90s.

Apparently because PS/2 mice were maxed out at about 200hz some people in the Quake 3 era would play at 125FPS-125Hz-125Hz(PS/2). 125 being the first of 3 major magic numbers with 250 and 333 being the other magic frame rate numbers.

Does it make sense maybe back in the 90s when information wasn't as widely available as it is now with testing. But within reasons having a higher polling rate as long as there is no crashing in certain scenarios is better and superior option. In other words the polling rate of the mice is more independent despite having the logic of locking things to 125 in the Quake 3 days.

It's not illogical it's just older information from an era when such testing and methodology were only nascent. I'd wager to state your 500-1000-2000-4000-8000Hz mouse is the superior way to go. Of course just because polling rates go that high doesn't mean all games take advantage of HD mice technologies nor all computers handle higher polling rates well. As seen in certain people they do state issues it does happen from time to time but it's there.

User avatar
Chief Blur Buster
Site Admin
Posts: 11648
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: Does matching polling rate to monitor Hz make sense or no?

Post by Chief Blur Buster » 06 Apr 2021, 17:09

With variable refresh rate (software-driven refresh cycles), it's possible to sync the monitor's refresh rate to the mouse poll rate, by synchronizing software-based game frame rate to a divisor of mouse poll rate. For example, 125fps, 250fps or 500fps.

However, oversampling (2000Hz-8000Hz) essentially also solves the problem as well, especially in the era of unsynchronized frame rates (aka VSYNC OFF). Where esports often use VSYNC OFF on fixed-Hz monitors, there is not as much sense in synchronizing poll rates to refresh rates.

For unsynchronized poll rates, you just need to oversample by about 6x, to prevent any mouse microstutters.

For example, these will be the sweet spots.
1000Hz+ poll rate for 144Hz-165Hz (past)
2000Hz+ poll rate for 240Hz-360Hz (now)
4000Hz+ poll rate for 480Hz-720Hz (~2025)

It is true that 1000Hz poll is too low for 360Hz monitors, since that's only a 3x oversample, when eliminating the mouse microstutters requires a higher oversampling factor, about 4 to 6x, preferably closer to at least 6x oversample.
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!

RocketRager
Posts: 21
Joined: 05 Apr 2021, 15:35

Re: Does matching polling rate to monitor Hz make sense or no?

Post by RocketRager » 06 Apr 2021, 17:11

Think it makes a difference because it’s controller joystick vs mouse? Probably not I guess.

Super frustrated with this damn game. Love it but hate it. I know what to do, and when I use solo play and can usually get into a nice flow with the ball and make some really nice touches.

As soon as competition enters the game with me, I cannot get to the ball as fast as them.

Is everyone just on 240-280 Hz monitors now and I’m the only one with 144? Feels like that just be the case.

RocketRager
Posts: 21
Joined: 05 Apr 2021, 15:35

Re: Does matching polling rate to monitor Hz make sense or no?

Post by RocketRager » 06 Apr 2021, 17:13

Chief Blur Buster wrote:
06 Apr 2021, 17:09
With variable refresh rate (software-driven refresh cycles), it's possible to sync the monitor's refresh rate to the mouse poll rate, by synchronizing software-based game frame rate to a divisor of mouse poll rate. For example, 125fps, 250fps or 500fps.

However, oversampling (2000Hz-8000Hz) essentially also solves the problem as well, especially in the era of unsynchronized frame rates (aka VSYNC OFF). Where esports often use VSYNC OFF on fixed-Hz monitors, there is not as much sense in synchronizing poll rates to refresh rates.

For unsynchronized poll rates, you just need to oversample by about 6x, to prevent any mouse microstutters.

For example, these will be the sweet spots.
1000Hz+ poll rate for 144Hz-165Hz (past)
2000Hz+ poll rate for 240Hz-360Hz (now)
4000Hz+ poll rate for 480Hz-720Hz (~2025)

It is true that 1000Hz poll is too low for 360Hz monitors, since that's only a 3x oversample, when eliminating the mouse microstutters requires a higher oversampling factor, about 4 to 6x, preferably closer to at least 6x oversample.
Sadly AFAIK these controllers only OC to 1k Hz. Best anyone on the planet can get for now.

Does that mean upgrading to 240-280-360 hz monitor is bad? If so, then I guess I have to quit the game lmao. Unless I go back to 60hz monitor ....hmmmm.

User avatar
Chief Blur Buster
Site Admin
Posts: 11648
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: Does matching polling rate to monitor Hz make sense or no?

Post by Chief Blur Buster » 06 Apr 2021, 17:31

RocketRager wrote:
06 Apr 2021, 17:13
Sadly AFAIK these controllers only OC to 1k Hz. Best anyone on the planet can get for now.

Does that mean upgrading to 240-280-360 hz monitor is bad? If so, then I guess I have to quit the game lmao. Unless I go back to 60hz monitor ....hmmmm.
No, it just reduces the upgradefeel versus theoretical maximum you can get.

A 240Hz monitor will still be a significant upgradefeel.

Here's an example: A theoretical perfect upgrade for 240Hz->360hz would be a 1.5x upgradefeel (perfect 360Hz 0ms GtG would have about 1/1.5th the motion blur of perfect 240Hz 0ms GtG, assuming no mouse microstutter bottlenecks). But the real world upgradefeel is closer to 1.1x to 1.2x from the dual combination of LCD GtG and the low mouse poll Hz. Upgrading to an 8000Hz mouse turns the 240Hz->360Hz into an ugpradefeel closer to 1.3x to 1.4x the 240Hz (assuming your frame rate can max out the 360 Hz monitor).

Now, 1000Hz is still fine for 240Hz monitors for most people. The upgrade from 144Hz to 240Hz is a 1.66x refresh rate upgrade (240fps at 240Hz getting clearer motion, at a ratio of 144/240ths the original motion blur of 144fps at 144Hz). But mouse microstutter vibrations re-adds a bit of motion blur (especially at high microstutter frequencies that are so fast, that the stutters blend to blur). You still keep most of the upgradefeel with just a 1000Hz mouse, think of it as probably being about a 1.4x upgradefeel at 1000Hz, and a 1.5x upgradefeel at 2000Hz (with the remainder being LCD GtG limitations). In other words, a smoother mouse feels like the framerate is higher. You're still getting most of the upgrade of 144->240Hz with just only a 4x poll Hz oversampling factor.

Eventually we need to invent tests that measures the motion blur degradation caused by high-frequency microstutter factors (such as the mouse). Few benchmarkers cover this territory. It is less of a problem at 240Hz but a bigger problem at 360Hz in reducing the magnitude of upgradefeel. I will probably be writing more articles about the stutter-to-blur continuum (stutters and blur are the same thing; it's just persistence at different frequencies -- like low-frequency vibrating music string that looks shaky -- versus high-frequency vibrating music string that looks blurry -- the best animation demo of the of the stutter-to-blur continuum is This variable-speed TestUFO Eye Tracking Animation but also TestUFO frame-rate ramping animations on VRR. The refresh rate and frame rate is one source of stutters, but mouse microstutters also contribute to the stutter-to-blur continuum problem which reduces the magnitude of upgradefeel if your poll rate is not sufficiently oversampled.

You can always upgrade your computer/GPU later.
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!

RocketRager
Posts: 21
Joined: 05 Apr 2021, 15:35

Re: Does matching polling rate to monitor Hz make sense or no?

Post by RocketRager » 06 Apr 2021, 17:42

Chief Blur Buster wrote:
06 Apr 2021, 17:31
RocketRager wrote:
06 Apr 2021, 17:13
Sadly AFAIK these controllers only OC to 1k Hz. Best anyone on the planet can get for now.

Does that mean upgrading to 240-280-360 hz monitor is bad? If so, then I guess I have to quit the game lmao. Unless I go back to 60hz monitor ....hmmmm.
No, it just reduces the upgradefeel versus theoretical maximum you can get.

A 240Hz monitor will still be a significant upgradefeel.

Here's an example: A theoretical perfect upgrade for 240Hz->360hz would be a 1.5x upgradefeel (perfect 360Hz 0ms GtG would have about 1/1.5th the motion blur of perfect 240Hz 0ms GtG, assuming no mouse microstutter bottlenecks). But the real world upgradefeel is closer to 1.1x to 1.2x from the dual combination of LCD GtG and the low mouse poll Hz. Upgrading to an 8000Hz mouse turns the 240Hz->360Hz into an ugpradefeel closer to 1.3x to 1.4x the 240Hz (assuming your frame rate can max out the 360 Hz monitor).

Now, 1000Hz is still fine for 240Hz monitors for most people. The upgrade from 144Hz to 240Hz is a 1.66x refresh rate upgrade (240fps at 240Hz getting clearer motion, at a ratio of 144/240ths the original motion blur of 144fps at 144Hz). But mouse microstutter vibrations re-adds a bit of motion blur (especially at high microstutter frequencies that are so fast, that the stutters blend to blur). You still keep most of the upgradefeel with just a 1000Hz mouse, think of it as probably being about a 1.4x upgradefeel at 1000Hz, and a 1.5x upgradefeel at 2000Hz (with the remainder being LCD GtG limitations). In other words, a smoother mouse feels like the framerate is higher. You're still getting most of the upgrade of 144->240Hz with just only a 4x poll Hz oversampling factor.

Eventually we need to invent tests that measures the motion blur degradation caused by high-frequency microstutter factors (such as the mouse). Few benchmarkers cover this territory. It is less of a problem at 240Hz but a bigger problem at 360Hz in reducing the magnitude of upgradefeel.

You can always upgrade your computer/GPU later.
God damn bro this is a lot of Greek but I think I’m following. Congrats on your site and knowledge. You should have a PhD in this shit from somewhere! Someone needs to grant you one.

I really need to know if it’s my AOC 144hz holding me back. Wife is gonna be pissed at me buying another monitor 6 months from the last one bit I have to know lmao!

Saw some dudes arguing back and forth about a BenQ vs an Asus in a different thread. I’ll probably try one of those I guess.

After that, if I don’t improve, maybe I’ll try internet investigation or just quit.

I have FTTH but unsure how the ISP “structures it”. It’s ATT so I assume it’s probably trash tier lol.

User avatar
Chief Blur Buster
Site Admin
Posts: 11648
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: Does matching polling rate to monitor Hz make sense or no?

Post by Chief Blur Buster » 06 Apr 2021, 17:45

I already teach some classrooms, sometimes, at vendors.
(I've quietly mentioned this publicly at services.blurbusters.com)

The easiest education -- about how stutters blend into motion blur -- is this animation:



Watch the moving TestUFO.
- At low frame rates, it vibrates a lot.
- But at high frame rates, the vibrations just blend into motion blur.

And you can look at the stationary UFO to show that it's perpetually just stationary vertical lines with horizontally-scrolling linegaps (At varying speed/framerate).

Consequently, this animation is almost a perfect demo of the stutter-to-blur continuum
- Like slow music strings (vibrates) versus fast music strings (blurs)
- Low framerate vibrates a lot, high framerate blurs (doubling framerate halves motion blur, as long as there's refresh rate available)

Similar issues also occurs to all sources of microstutters -- from unsynchronized frequencies (framerate-vs-refreshrate, framerate-vs-pollrate, etc). If something microstutters rapidly 2 pixels 70 times a second, that turns into a 2-pixel blur instead of 2-pixel vibration. Oversampling helps quite a lot (e.g. 73fps stutters a lot less at 360Hz than at 144Hz..... Likewise also the inverse too, doing VSYNC OFF 500fps at 144Hz stutters quite a lot less than plain old VSYNC OFF 147fps at 144Hz). Using VRR instead of VSYNC OFF solves the Hz-vs-framerate sync problem, but still leaves the framerate-vs-pollrate unsolved and still requires heavy oversampling to solve for now...
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!

User avatar
Chief Blur Buster
Site Admin
Posts: 11648
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: Does matching polling rate to monitor Hz make sense or no?

Post by Chief Blur Buster » 06 Apr 2021, 17:48

Theoretically Possible: Custom Frame Rate Capping Software That Syncs To Poll Rate
If you are a software developer, there is a solution
Now, there are potential workarounds, like writing a custom frame-rate capper that uses a mousehook (LowLevelMouseProc callback and the WH_MOUSE_LL event). Basically an RTSS clone that monitors the mouse poll rate and synchronizes frame rate to a divisor of the mouse poll rate. And using a 280 Hz VRR monitor to achieve the low-lag 250fps frame rate cap synchronized to poll rate (1000Hz).

Recipie:
1. VRR display
2. Refresh rate capable of syncing to poll rate
3. Custom frame rate capping software that syncs to poll rate (or divisor thereof)

However, just merely upgrading to 240Hz display refresh rate will still be a big upgrade even for refresh-rate-unsynchronized 1000Hz device poll rate.
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!

RocketRager
Posts: 21
Joined: 05 Apr 2021, 15:35

Re: Does matching polling rate to monitor Hz make sense or no?

Post by RocketRager » 06 Apr 2021, 17:55

Chief Blur Buster wrote:
06 Apr 2021, 17:48
Theoretically Possible: Custom Frame Rate Capping Software That Syncs To Poll Rate
If you are a software developer, there is a solution
Now, there are potential workarounds, like writing a custom frame-rate capper that uses a mousehook (LowLevelMouseProc callback and the WH_MOUSE_LL event). Basically an RTSS clone that monitors the mouse poll rate and synchronizes frame rate to a divisor of the mouse poll rate. And using a 280 Hz VRR monitor to achieve the low-lag 250fps frame rate cap synchronized to poll rate (1000Hz).

Recipie:
1. VRR display
2. Refresh rate capable of syncing to poll rate
3. Custom frame rate capping software that syncs to poll rate (or divisor thereof)

However, just merely upgrading to 240Hz display refresh rate will still be a big upgrade even for refresh-rate-unsynchronized 1000Hz device poll rate.
A real hero. If I rank up past champ, I’m donating to the site somehow lol ;)

Post Reply