The main reason for the 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.
MegaMelmek
Posts: 235
Joined: 21 Jan 2021, 12:54

Re: The main reason for the input lag.

Post by MegaMelmek » 07 Dec 2021, 17:16

a_c_r_e_a_l wrote:
06 Dec 2021, 18:16
MegaMelmek wrote:
06 Dec 2021, 14:01
Betwen ground and zero i have 0V. Freq is all right. I spend crazy money, change meter, no harmonics so i give up mate. Mouse i floaty no hit reg… 3 diferent ISP fiber VDSL, no fix, 5 diferent PC. There is no solution boys.
You need more than an ordinary multimeter. I have 0.05V between ground and neutral, stable voltage at all times, 50Hz ± 0.09. What you need is to check wirings for radio frequency interference. Strong transceivers nearby your house/flat can lead to induction of interference in the electrical wiring. Find the radio specialist in your town and go ahead.
Well mate it was measured with 400Euro meter so there is no false number betwen ground and zero is constantly 0V. All my wire in my appartment is coper. There was one concern that i made with half answer. That i put everywhere coax cable in the wall for TV. There was measured 40V between ground and coax cable and that was wierd to me. That coax cable is not conected to anything so double wierd. A was asking why is that hapening what volatage that is and the answer was: inducted voltage the coax cable is close to elctrical cable its charging by itself it does not mean if you want light up a bulb with that 40V it never happend. The expensive meter was able to measure 12V between ground and my table or metal chair. But it never measure any voltage between ground and zero. If you have measured voltage here maybe you have diffrent issue and if you measure voltage between ground and zero you groud is not working as it have too.
Attachments
IMG_9624.jpg
IMG_9624.jpg (1018.89 KiB) Viewed 7163 times

woodyfly
Posts: 91
Joined: 03 Jul 2020, 07:53

Re: The main reason for the input lag.

Post by woodyfly » 08 Dec 2021, 11:24

what the fuck did u just say.

Qure
Posts: 5
Joined: 11 Feb 2020, 03:01

Re: The main reason for the input lag.

Post by Qure » 08 Jan 2022, 23:50

MegaMelmek wrote:
07 Dec 2021, 17:16
a_c_r_e_a_l wrote:
06 Dec 2021, 18:16
MegaMelmek wrote:
06 Dec 2021, 14:01
Betwen ground and zero i have 0V. Freq is all right. I spend crazy money, change meter, no harmonics so i give up mate. Mouse i floaty no hit reg… 3 diferent ISP fiber VDSL, no fix, 5 diferent PC. There is no solution boys.
You need more than an ordinary multimeter. I have 0.05V between ground and neutral, stable voltage at all times, 50Hz ± 0.09. What you need is to check wirings for radio frequency interference. Strong transceivers nearby your house/flat can lead to induction of interference in the electrical wiring. Find the radio specialist in your town and go ahead.
Well mate it was measured with 400Euro meter so there is no false number betwen ground and zero is constantly 0V. All my wire in my appartment is coper. There was one concern that i made with half answer. That i put everywhere coax cable in the wall for TV. There was measured 40V between ground and coax cable and that was wierd to me. That coax cable is not conected to anything so double wierd. A was asking why is that hapening what volatage that is and the answer was: inducted voltage the coax cable is close to elctrical cable its charging by itself it does not mean if you want light up a bulb with that 40V it never happend. The expensive meter was able to measure 12V between ground and my table or metal chair. But it never measure any voltage between ground and zero. If you have measured voltage here maybe you have diffrent issue and if you measure voltage between ground and zero you groud is not working as it have too.
Have you tested if your USB Hub is the culprit?

You might think it's silly, but I've found out myself that I have 2 USB Hubs that give horrible Mouse Tester results for ALL my Mouses. (I have 4 mouses)
Only my third USB Hub gives me really clean result for the '1000mhz' test, the other USB Hubs will flactuate horribily on every test.

Not only that, but you can mess around with 'Interrupt Affinity Policy Tool' to assign your USB Hub to a different CPU Core to clear your Mouse & Keyboard interrupts. (Just be careful with this tool since if used incorrectly it can damage certain hardware)
Windows usually puts most of the Interrupts on the first CPU Core, so if you assign your USB Hub to a diffferent one, it can make your Mouse feel "light" as a feather all of sudden. (I've noticed it a clear difference while I was aim-labbing on CS:GO workshop maps.)
Also, there are specific 'CPU Cores' that will give better responsiveness "feeling" for your Mouse/Keyboard, while other CPU Cores will give worse results. So you have to test every core.
I suggest testing it while you aim-lab to see the difference in an actualy game and not use the Desktop to test this. (My best CPU Core for my Mouse/Keyboard was 'CPU 02' which is CPU Core 3. They count 'CPU 00' as your First Core. My CPU is Ryzen 3900x)

Slender
Posts: 573
Joined: 25 Jan 2020, 17:55

Re: The main reason for the input lag.

Post by Slender » 09 Jan 2022, 06:16

i have z390 asus tuf, it is normal what after clear cmos, profiles in bios save?

InputLagger
Posts: 198
Joined: 13 Sep 2021, 12:39
Location: RUS

Re: The main reason for the input lag.

Post by InputLagger » 09 Jan 2022, 14:48

Slender wrote:
09 Jan 2022, 06:16
i have z390 asus tuf, it is normal what after clear cmos, profiles in bios save?
Yep, I have PC setup from 2013-14 years Am3+, and I did a little research for this, some days ago and only solution for this will be: reflash bios completely :roll: . Idk why vedors still don't add removal feature

Slender
Posts: 573
Joined: 25 Jan 2020, 17:55

Re: The main reason for the input lag.

Post by Slender » 09 Jan 2022, 22:05

InputLagger wrote:
09 Jan 2022, 14:48
Slender wrote:
09 Jan 2022, 06:16
i have z390 asus tuf, it is normal what after clear cmos, profiles in bios save?
Yep, I have PC setup from 2013-14 years Am3+, and I did a little research for this, some days ago and only solution for this will be: reflash bios completely :roll: . Idk why vedors still don't add removal feature
after bios updating, profiles saving.

MegaMelmek
Posts: 235
Joined: 21 Jan 2021, 12:54

Re: The main reason for the input lag.

Post by MegaMelmek » 16 Jan 2022, 08:06

Qure wrote:
08 Jan 2022, 23:50
MegaMelmek wrote:
07 Dec 2021, 17:16
a_c_r_e_a_l wrote:
06 Dec 2021, 18:16
MegaMelmek wrote:
06 Dec 2021, 14:01
Betwen ground and zero i have 0V. Freq is all right. I spend crazy money, change meter, no harmonics so i give up mate. Mouse i floaty no hit reg… 3 diferent ISP fiber VDSL, no fix, 5 diferent PC. There is no solution boys.
You need more than an ordinary multimeter. I have 0.05V between ground and neutral, stable voltage at all times, 50Hz ± 0.09. What you need is to check wirings for radio frequency interference. Strong transceivers nearby your house/flat can lead to induction of interference in the electrical wiring. Find the radio specialist in your town and go ahead.
Well mate it was measured with 400Euro meter so there is no false number betwen ground and zero is constantly 0V. All my wire in my appartment is coper. There was one concern that i made with half answer. That i put everywhere coax cable in the wall for TV. There was measured 40V between ground and coax cable and that was wierd to me. That coax cable is not conected to anything so double wierd. A was asking why is that hapening what volatage that is and the answer was: inducted voltage the coax cable is close to elctrical cable its charging by itself it does not mean if you want light up a bulb with that 40V it never happend. The expensive meter was able to measure 12V between ground and my table or metal chair. But it never measure any voltage between ground and zero. If you have measured voltage here maybe you have diffrent issue and if you measure voltage between ground and zero you groud is not working as it have too.
Have you tested if your USB Hub is the culprit?

You might think it's silly, but I've found out myself that I have 2 USB Hubs that give horrible Mouse Tester results for ALL my Mouses. (I have 4 mouses)
Only my third USB Hub gives me really clean result for the '1000mhz' test, the other USB Hubs will flactuate horribily on every test.

Not only that, but you can mess around with 'Interrupt Affinity Policy Tool' to assign your USB Hub to a different CPU Core to clear your Mouse & Keyboard interrupts. (Just be careful with this tool since if used incorrectly it can damage certain hardware)
Windows usually puts most of the Interrupts on the first CPU Core, so if you assign your USB Hub to a diffferent one, it can make your Mouse feel "light" as a feather all of sudden. (I've noticed it a clear difference while I was aim-labbing on CS:GO workshop maps.)
Also, there are specific 'CPU Cores' that will give better responsiveness "feeling" for your Mouse/Keyboard, while other CPU Cores will give worse results. So you have to test every core.
I suggest testing it while you aim-lab to see the difference in an actualy game and not use the Desktop to test this. (My best CPU Core for my Mouse/Keyboard was 'CPU 02' which is CPU Core 3. They count 'CPU 00' as your First Core. My CPU is Ryzen 3900x)
And how i tell my pc that i want that core no3 will only operate mouse?

MatrixQW
Posts: 278
Joined: 07 Jan 2019, 10:01

Re: The main reason for the input lag.

Post by MatrixQW » 17 Jan 2022, 17:29

MegaMelmek wrote:
16 Jan 2022, 08:06
Qure wrote:
08 Jan 2022, 23:50
Have you tested if your USB Hub is the culprit?

You might think it's silly, but I've found out myself that I have 2 USB Hubs that give horrible Mouse Tester results for ALL my Mouses. (I have 4 mouses)
Only my third USB Hub gives me really clean result for the '1000mhz' test, the other USB Hubs will flactuate horribily on every test.

Not only that, but you can mess around with 'Interrupt Affinity Policy Tool' to assign your USB Hub to a different CPU Core to clear your Mouse & Keyboard interrupts. (Just be careful with this tool since if used incorrectly it can damage certain hardware)
Windows usually puts most of the Interrupts on the first CPU Core, so if you assign your USB Hub to a diffferent one, it can make your Mouse feel "light" as a feather all of sudden. (I've noticed it a clear difference while I was aim-labbing on CS:GO workshop maps.)
Also, there are specific 'CPU Cores' that will give better responsiveness "feeling" for your Mouse/Keyboard, while other CPU Cores will give worse results. So you have to test every core.
I suggest testing it while you aim-lab to see the difference in an actualy game and not use the Desktop to test this. (My best CPU Core for my Mouse/Keyboard was 'CPU 02' which is CPU Core 3. They count 'CPU 00' as your First Core. My CPU is Ryzen 3900x)
And how i tell my pc that i want that core no3 will only operate mouse?
Interrupt affinity only reduces ISR/DPC and mostly changing GPU.
You are not going to get reduced input lag or more fps.
I tried this long ago and never had any benefits.
It could generate other oddities.
If a game's graphics thread gets set on cpu 3 and you set interrupts there also (GPU, mouse, whatever) it could actually make things worse.


The app let's you specify.

To see if it's working install "Windows Performance Toolkit" and run cmd prompt with admin rights.

1. cd "C:\Program Files (x86)\Windows Kits\10\Windows Performance Toolkit"

2. xperf -on base+interrupt+dpc (move mouse really fast on desktop for 8 seconds or so)

3. xperf -d C:\Users\YourName\Documents\trace.etl

4. xperf -i C:\Users\YourName\Documents\trace.etl -o C:\Users\YourName\Documents\report.txt -a dpcisr

Now check the DPC/Interrupt for usbport.sys in report.txt to see if the specified cpu has a large number in 'usec'.
.
.
affinity.jpg
affinity.jpg (144.17 KiB) Viewed 6189 times

MegaMelmek
Posts: 235
Joined: 21 Jan 2021, 12:54

Re: The main reason for the input lag.

Post by MegaMelmek » 17 Jan 2022, 18:14

MatrixQW wrote:
17 Jan 2022, 17:29
MegaMelmek wrote:
16 Jan 2022, 08:06
Qure wrote:
08 Jan 2022, 23:50
Have you tested if your USB Hub is the culprit?

You might think it's silly, but I've found out myself that I have 2 USB Hubs that give horrible Mouse Tester results for ALL my Mouses. (I have 4 mouses)
Only my third USB Hub gives me really clean result for the '1000mhz' test, the other USB Hubs will flactuate horribily on every test.

Not only that, but you can mess around with 'Interrupt Affinity Policy Tool' to assign your USB Hub to a different CPU Core to clear your Mouse & Keyboard interrupts. (Just be careful with this tool since if used incorrectly it can damage certain hardware)
Windows usually puts most of the Interrupts on the first CPU Core, so if you assign your USB Hub to a diffferent one, it can make your Mouse feel "light" as a feather all of sudden. (I've noticed it a clear difference while I was aim-labbing on CS:GO workshop maps.)
Also, there are specific 'CPU Cores' that will give better responsiveness "feeling" for your Mouse/Keyboard, while other CPU Cores will give worse results. So you have to test every core.
I suggest testing it while you aim-lab to see the difference in an actualy game and not use the Desktop to test this. (My best CPU Core for my Mouse/Keyboard was 'CPU 02' which is CPU Core 3. They count 'CPU 00' as your First Core. My CPU is Ryzen 3900x)
And how i tell my pc that i want that core no3 will only operate mouse?
Interrupt affinity only reduces ISR/DPC and mostly changing GPU.
You are not going to get reduced input lag or more fps.
I tried this long ago and never had any benefits.
It could generate other oddities.
If a game's graphics thread gets set on cpu 3 and you set interrupts there also (GPU, mouse, whatever) it could actually make things worse.


The app let's you specify.

To see if it's working install "Windows Performance Toolkit" and run cmd prompt with admin rights.

1. cd "C:\Program Files (x86)\Windows Kits\10\Windows Performance Toolkit"

2. xperf -on base+interrupt+dpc (move mouse really fast on desktop for 8 seconds or so)

3. xperf -d C:\Users\YourName\Documents\trace.etl

4. xperf -i C:\Users\YourName\Documents\trace.etl -o C:\Users\YourName\Documents\report.txt -a dpcisr

Now check the DPC/Interrupt for usbport.sys in report.txt to see if the specified cpu has a large number in 'usec'.
.
.
affinity.jpg
I realy belive in your tweeks (they make sense to me) and i can say i found improve in them with my input lag. (two days in pretty smooth gaming.. i play only CS GO)
Can i ask what usb hub you use (i mean the brand)?
The core tweek i will try but i am on windows 11 hope it will be same, also i can say allready my DPC all pretty low compare what they was 1 year back on w10….
Can confirm that this insane input lag has to do nothing with internet provider and i am 101% sure of that

MatrixQW
Posts: 278
Joined: 07 Jan 2019, 10:01

Re: The main reason for the input lag.

Post by MatrixQW » 17 Jan 2022, 18:35

I think you are confusing me with the user Qure.
I'm not using an external hub, it's the one from my board Asus Z97-K but connected to a 2.0 port.
That's why it is using the 'enhanced pci to usb host controller'.

Post Reply