Optimization Hub for beginners

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

Re: Optimization Hub for beginners

Post by Chief Blur Buster » 30 Dec 2020, 21:18

Yep, the thread I created -- linkified for other readers. ;)

Even sub-millisecond latency improvements can still add up for elite esports atheletes -- shaving off all those 0.5ms-increasers at all stages of the latency chain, improved mouse, improved display, improved refresh rates, etc.

However, just wanted to generalize, that statistically for the majority of average population, the higher poll rates benefits is far more noticeable fluidity-side rather than latency-side. So I nuanced it with "most" -- still acknowledging the latency benefits that can contribute.

But having both is a bonus -- having cake and eating it too!

A testing caveat though: One big part of the lag decrease of the Viper mouse is the reduced latency of the optical mouse switches. I can actually feel that my Razer 8Khz Viper's optically-actuated mouse buttons "feel faster" -- they are noticeably faster-behaving than the mechanical-contact mouse buttons of most other gaming mice. It's the most tactile-trustfully-clicky mouse I've ever used; that's the optically actuated switches! The design of which also eliminate a lot of need for antibounce latencies. The improvements in clickfeel even still occur when I intentionally reduce poll rate to 1000 Hz or 2000 Hz on the Razer 8Khz Viper. This is partially why for a more apples-vs-apples lagfeel comparison, you have to compare 1000Hz-vs-8000Hz on the same mouse (the Razer 8KHz Viper) to isolate pollrate-related lagfeel improvements, since most of the button lag improvement of the Viper mouse is the faster human actuation of the optical mouse buttons.

Thanks to the optical mouse button switches, it almost feels like the mouse is clicking ahead of itself when I click quickly on the Razer Viper 8KHz -- when I go back to any mechanical-actuated-button mice, I can actually tell the clickfeel difference. I'm not even an esports player though I'm an important person in the advocacy of the technology used by esports athletes. Yet, I can feel the click latency difference even when just using a mere web browser when clicking things in a hurry. There's enough milliseconds of lag savings clearly for me to notice the clickfeel improvements in plain regular Windows use, slight as it may be. Versus other Razer 1000 Hz mice, Logitech 1000 Hz mice, and ASUS 1000 Hz mice (all mechanical-actuated-button).

Combined latency decrease from increased pollrate AND the optical switch AND other optimizations that Razer may have done (e.g. less need for antibounce lag), definitely seem to build up human-noticeably in mere Windows use for me. Somehow my button feels like they're perfectly predicting slightly ahead of my actual clicks. Never had mouse buttons feel this responsive when clicking on local UI's like pulldown menus, buttons, and other Windows GUI elements. Measuring this reliably is another question.

Past tests with mouse lag involved attaching a LED to the mouse button, but that doesn't measure latency before actuation (e.g. click actuation of the switch), so doing the same LED test won't reveal the optical switch speedup, since actuation/pressure/click lag is before LED illumination.

A new test from the first touch to first-actuation would be needed to measure the latency decreases. But I would not be surprised to see a decrease of a few milliseconds with a different lag test from "zero finger pressure" to "mouse event triggered".

I am not surprised that diakou has noticed the benefits...
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
xeos
Posts: 43
Joined: 12 Jul 2018, 14:54
Contact:

Re: Optimization Hub for beginners

Post by xeos » 01 Jan 2021, 11:15

Chief Blur Buster wrote:
29 Dec 2020, 22:56

(2) Tests from an Arduino mouse-click simulator from 125Hz, 500Hz, and 1000Hz, show that click latency generally decreases at higher poll rates, assuming same-device poll-rate changes, at least on uncongested USB ports (no competition with other USB traffic). The Arduino-compatible simulates a click, and a photodiode attached to the same Arduino-compatible microcontroller. Connected back to the same program running on the Ardunio. So synthetic data results are made possible because the same Arduino-comaptible program is configuring the poll rate, and then sending the simulated USB HID mouse click (lag stopwatch start) and measuring for the photodiode (lag stopwatch end). Results are fully consistent with the math formula in (1) within error margins.
where's the link? I'd like to read about that setup and results in more detail.
Measure display input lag the cheap way or the best way (IMHO, but I'm biased).

andrelip
Posts: 160
Joined: 21 Mar 2014, 17:50

Re: Optimization Hub for beginners

Post by andrelip » 02 Jan 2021, 11:13

xeos wrote:
10 Dec 2020, 00:37
Some of the stuff seems pretty hokey but hey, an open minded person who measures the results empirically will learn more than a closed minded person. But to do that you need actual objective tools, not "feel".

For instance, one of your links is to custom resolutions (https://www.monitortests.com/forum/Thre ... tility-CRU). I'd be very wary of that - monitors often will accept timing patterns that they are not designed for, but do bad things with them (read, buffer them and display them under supported timing with extra lag added). I'd suggest linking to the more affordable monitor lag testing tools out there:

piLagtester (~free)
piLagTesterPRO ($40)
Time Sleuth ($75).

Disclaimer: I'm the author of the first two (see my sig for links).
Thank you so much! I reduced my input lag by more than 40ms

Context:

I'm traveling to my parent's house and using my 16 MacBook PRO and LG C9 (OLED) as a monitor.
This TV is reported to have low latency and the Macbook PRO has high performance after the VRM mod, but I was sucking in the game.
Thought it was energy-saving, latency from the USB-C->HDMI adapter, and many other things and the problem was exactly what you told about timing buffer.

How to solve:

I'm using display scaling since I don't have HDMI 2.1 to use the native 4k 120hz. I checked the timings from the default resolutions using CRU and then discovered with the pattern it was (automatic PC in my case). Then I created low resolutions @ 120hz using that pattern instead of the ones generated by the AMD driver.

I couldn't even score 190 ms in Humanbenchmark and now I did one average of 146ms in 5 runs. I never did this average before with AW2518hf (240hz) and PG259QN (360hz). Amazing!

Conclusions

I don't know if this is due to OLED having better response time or wrong timing in the others. I will probably Bruteforce my home monitor (PG259QN) with every pattern to see if I can find another magic one that reduces signal processing.

It’s ridiculous how much latency it adds to support compatibility with other modes. I always expected the timing to generate diminished returns like small gradient changes or maybe a full frame but never 40ms (almost 5 frames at 120hz).

Nobody is even talking about it. In many years all I read was the bug of 60hz in the first generation of 240hz display (and that is probably the solution). Now I am guessing if some overclocks and large VT really worth it, if the RLA would detect this type of issue and many other things...

Can't thank you enough

Brainlet
Posts: 100
Joined: 30 May 2020, 12:39
Contact:

Re: Optimization Hub for beginners

Post by Brainlet » 03 Jan 2021, 15:11

Similar to diakou's example of snowballing, NVIDIA also conducted a research about how minor latencies can snowball into something much bigger. I added it to this infograph:

Image
Starting point for beginners: PC Optimization Hub

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

Re: Optimization Hub for beginners

Post by Chief Blur Buster » 03 Jan 2021, 18:46

Brainlet wrote:
03 Jan 2021, 15:11
Similar to diakou's example of snowballing, NVIDIA also conducted a research about how minor latencies can snowball into something much bigger. I added it to this infograph:
Another factor is latency volatility (and its related cousin: erratic stutter). Imagine true erratic randomization of latency (and continually varying amplitude of randomization).

Small amounts of latency volatility like that can throw gaming off even in the single-digit millisecond timescales nowadays with the Vicious Cycle Effect (higher Hz + higher resolution = ever tinier imperfections visible).
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
xeos
Posts: 43
Joined: 12 Jul 2018, 14:54
Contact:

Re: Optimization Hub for beginners

Post by xeos » 03 Jan 2021, 19:30

Chief Blur Buster wrote:
03 Jan 2021, 18:46
Brainlet wrote:
03 Jan 2021, 15:11
Similar to diakou's example of snowballing, NVIDIA also conducted a research about how minor latencies can snowball into something much bigger. I added it to this infograph:
Another factor is latency volatility (and its related cousin: erratic stutter). Imagine true erratic randomization of latency (and continually varying amplitude of randomization).
I'd go so far as to suggest that 120hz locked in might be preferable to unlocked frame rates bouncing around between 120hz and 240hz. At the very least the question needs to be asked empirically.
Measure display input lag the cheap way or the best way (IMHO, but I'm biased).

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

Re: Optimization Hub for beginners

Post by Chief Blur Buster » 11 Feb 2021, 19:16

xeos wrote:
03 Jan 2021, 19:30
Chief Blur Buster wrote:
03 Jan 2021, 18:46
Brainlet wrote:
03 Jan 2021, 15:11
Similar to diakou's example of snowballing, NVIDIA also conducted a research about how minor latencies can snowball into something much bigger. I added it to this infograph:
Another factor is latency volatility (and its related cousin: erratic stutter). Imagine true erratic randomization of latency (and continually varying amplitude of randomization).
I'd go so far as to suggest that 120hz locked in might be preferable to unlocked frame rates bouncing around between 120hz and 240hz. At the very least the question needs to be asked empirically.
In certain cases, it can certainly be. Enabling VRR and capping at middle of VRR range also improves latency consistency. While it is not as low lag as VSYNC OFF framerate above refresh rates but if you've got a great max-Hz (240Hz, 280Hz or 360Hz), capping at a specific frame rate generates delicious consistent latency with much, much less lag than VSYNC ON.

VRR is the world's lowest "non-VSYNC-OFF" sync technology.
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!

NDUS
Posts: 71
Joined: 12 Aug 2019, 16:05

Re: Optimization Hub for beginners

Post by NDUS » 11 Feb 2021, 20:21

Just want to add that https://github.com/miniant-git/REAL is a way to reduce a little bit of audio latency in some systems.

I also performed some audio latency testing on the following devices, in CSGO, using a 400FPS cap and the minimum allowable audio buffer

Gigabyte Z390 Aorus Pro onboard audio: ~69ms ±15
Creative Sound Blaster AE-9 direct mode on: ~71ms ±15
Creative Sound Blaster AE-9 direct mode off: ~100ms ±25
Topping E30: ~104ms ±30

slaver01
Posts: 89
Joined: 21 Sep 2020, 01:48

Re: Optimization Hub for beginners

Post by slaver01 » 10 Mar 2021, 21:31

Chief Blur Buster wrote:
30 Dec 2020, 21:18
Yep, the thread I created -- linkified for other readers. ;)

Even sub-millisecond latency improvements can still add up for elite esports atheletes -- shaving off all those 0.5ms-increasers at all stages of the latency chain, improved mouse, improved display, improved refresh rates, etc.

However, just wanted to generalize, that statistically for the majority of average population, the higher poll rates benefits is far more noticeable fluidity-side rather than latency-side. So I nuanced it with "most" -- still acknowledging the latency benefits that can contribute.

But having both is a bonus -- having cake and eating it too!

A testing caveat though: One big part of the lag decrease of the Viper mouse is the reduced latency of the optical mouse switches. I can actually feel that my Razer 8Khz Viper's optically-actuated mouse buttons "feel faster" -- they are noticeably faster-behaving than the mechanical-contact mouse buttons of most other gaming mice. It's the most tactile-trustfully-clicky mouse I've ever used; that's the optically actuated switches! The design of which also eliminate a lot of need for antibounce latencies. The improvements in clickfeel even still occur when I intentionally reduce poll rate to 1000 Hz or 2000 Hz on the Razer 8Khz Viper. This is partially why for a more apples-vs-apples lagfeel comparison, you have to compare 1000Hz-vs-8000Hz on the same mouse (the Razer 8KHz Viper) to isolate pollrate-related lagfeel improvements, since most of the button lag improvement of the Viper mouse is the faster human actuation of the optical mouse buttons.

Thanks to the optical mouse button switches, it almost feels like the mouse is clicking ahead of itself when I click quickly on the Razer Viper 8KHz -- when I go back to any mechanical-actuated-button mice, I can actually tell the clickfeel difference. I'm not even an esports player though I'm an important person in the advocacy of the technology used by esports athletes. Yet, I can feel the click latency difference even when just using a mere web browser when clicking things in a hurry. There's enough milliseconds of lag savings clearly for me to notice the clickfeel improvements in plain regular Windows use, slight as it may be. Versus other Razer 1000 Hz mice, Logitech 1000 Hz mice, and ASUS 1000 Hz mice (all mechanical-actuated-button).

Combined latency decrease from increased pollrate AND the optical switch AND other optimizations that Razer may have done (e.g. less need for antibounce lag), definitely seem to build up human-noticeably in mere Windows use for me. Somehow my button feels like they're perfectly predicting slightly ahead of my actual clicks. Never had mouse buttons feel this responsive when clicking on local UI's like pulldown menus, buttons, and other Windows GUI elements. Measuring this reliably is another question.

Past tests with mouse lag involved attaching a LED to the mouse button, but that doesn't measure latency before actuation (e.g. click actuation of the switch), so doing the same LED test won't reveal the optical switch speedup, since actuation/pressure/click lag is before LED illumination.

A new test from the first touch to first-actuation would be needed to measure the latency decreases. But I would not be surprised to see a decrease of a few milliseconds with a different lag test from "zero finger pressure" to "mouse event triggered".

I am not surprised that diakou has noticed the benefits...
are you saying not to use "Timer Resolution" * ISLC *? Will it benefit more?

empleat
Posts: 149
Joined: 28 Feb 2020, 21:06

Re: Optimization Hub for beginners

Post by empleat » 02 Aug 2021, 18:08

Hahaha I like how you called article boring boredom :D :D :D Beginners? This is actually pretty good list containing advanced stuff :lol: Last time I read basic tweaks, google didn't even find them :D I like idea having separate system for gaming, but holly shit, every time I need something to restart PC? Even on NVME and Intel mobo it takes 10 seconds and fast boot is crap! I have 2 computers, but old is too noisy, also since I built my new computer to be silent as fly even in max load!

I linked your guide in my 70k view tweakguide, if it is okay. So people see this, you complied pretty good list :D

Post Reply