I have the new Razer 8000 Hz prototype gaming mouse on my desk.

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.
Locked
User avatar
lyrill
Posts: 385
Joined: 06 Oct 2020, 10:37

Re: I have the new Razer 8000 Hz prototype gaming mouse on my desk.

Post by lyrill » 02 Nov 2020, 10:49

O_O wrote:
30 Oct 2020, 21:39
howiec wrote:
27 Oct 2020, 14:23

I'm curious if anyone has access to mouse firmware source code for any of the Pixart sensors, e.g. 3389 and can legally share it.
I'm not familiar with the underlying math and physics that factors in LOD values (clearly we're likely talking some optics too).
If you look at the datasheet for the ADNS-3050 it has some code for LOD although being an entry level sensor no fancy registers, just works of surface quality and shutter value IIRC. I guess the good news it that it wouldn't affect CPI :D other than that probably not really what your looking for.

which sensors apart from 3050 adjust LOD without affecting dpi?

if it's SQUAL related shouldn't it be pretty much all mouse that has a SQUAL feature? like the roccat and mionix stuff from 2012. those were avago 9800 btw.






also woah 15 years since 1khz. about damn time to upgrade.

User avatar
Chief Blur Buster
Site Admin
Posts: 11653
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: I have the new Razer 8000 Hz prototype gaming mouse on my desk.

Post by Chief Blur Buster » 02 Nov 2020, 16:37

lyrill wrote:
02 Nov 2020, 10:49
also woah 15 years since 1khz. about damn time to upgrade.
Yep, for 360Hz monitors, you need at least 2000 Hz poll rate to improve the ability to tell the difference between 240Hz vs 360Hz.

Fortunately going past 1000Hz means 8000Hz is a freebie feature due to USB2 protocol. You can select 125Hz, 500Hz, 1000Hz, 2000Hz, 4000Hz and 8000Hz.
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!

O_O
Posts: 19
Joined: 26 Oct 2020, 06:19

Re: I have the new Razer 8000 Hz prototype gaming mouse on my desk.

Post by O_O » 02 Nov 2020, 20:06

lyrill wrote:
02 Nov 2020, 10:49
which sensors apart from 3050 adjust LOD without affecting dpi?

if it's SQUAL related shouldn't it be pretty much all mouse that has a SQUAL feature? like the roccat and mionix stuff from 2012. those were avago 9800 btw.
Specifically SQual and Shutter are read only registers so since they cannot be written to then there's no affect to the sensor operation. The more advanced sensors may have several writable registers pertaining to LOD and therefore affect the sensor.

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

Re: I have the new Razer 8000 Hz prototype gaming mouse on my desk.

Post by slaver01 » 03 Nov 2020, 03:48

Chief Blur Buster wrote:
02 Nov 2020, 16:37
lyrill wrote:
02 Nov 2020, 10:49
also woah 15 years since 1khz. about damn time to upgrade.
Yep, for 360Hz monitors, you need at least 2000 Hz poll rate to improve the ability to tell the difference between 240Hz vs 360Hz.

Fortunately going past 1000Hz means 8000Hz is a freebie feature due to USB2 protocol. You can select 125Hz, 500Hz, 1000Hz, 2000Hz, 4000Hz and 8000Hz.
Playing a Fortnite with G403 Hero mouse and 240hz monitor could I benefit from 2000 hz? Dpi? How can I overclock my G403?

User avatar
lyrill
Posts: 385
Joined: 06 Oct 2020, 10:37

Re: I have the new Razer 8000 Hz prototype gaming mouse on my desk.

Post by lyrill » 03 Nov 2020, 11:31

Chief Blur Buster wrote:
02 Nov 2020, 16:37
lyrill wrote:
02 Nov 2020, 10:49
also woah 15 years since 1khz. about damn time to upgrade.
Yep, for 360Hz monitors, you need at least 2000 Hz poll rate to improve the ability to tell the difference between 240Hz vs 360Hz.

Fortunately going past 1000Hz means 8000Hz is a freebie feature due to USB2 protocol. You can select 125Hz, 500Hz, 1000Hz, 2000Hz, 4000Hz and 8000Hz.
yeah and also 8k+fps mouse sensors also been around about as long

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

Re: I have the new Razer 8000 Hz prototype gaming mouse on my desk.

Post by empleat » 04 Nov 2020, 11:02

Wow i heard about 2k, but didn't expect 8k :D Now with nvidia reflex and 360hz, there is no better time now, than ever to enjoy competitive games. But i have repetitive strain injury, that sucks... I was using 500hz/800dpi and i was supreme master class in cs go. Because with 1000hz, it captures even smallest movements and even aim is more snappier, it is harder to do smaller adjustments. Also timer resolution is important with this, if you use 500hz polling on mouse, you may not want to use 0.5ms timer. Because than movement is more inconsistent, as sometimes last position is reported and sometimes last is missed. Than position up to 4ms in past gets rendered, but than you may capture one almost instantly, because frequent updates to the cpu from resolution timer.

1000hz is kinda meh, 2k probably too, because inconsistency. But 8k hmm?! Pros use low dpi anyways, under 800. It may be time to lower your sensitivity to like 400-600 dpi :D Still i would be kinda worried, because it will render smallest movements!

When you would have 8k (0.125ms) polling rate, than even if you miss one polling period, it will render previous one in only up to 0.250ms in past. So i can imagine, that it will be smoother. Low DPC. latency motherboards will be absolutely crucial for this!!! Right now: there is only minimum of motherboards, which have DPC latency below 100 us, yet it can differ during system load and per hw configurations!!! Only anandtech is currently doing DPC latency tests, because they are relative and differ per hardware configurations, even main source of DPC latency are badly coded drivers and not the hardware.

Even 500$ asus motherboards suffer even from 1ms DPC latency :D This will be brutal, people with shitty motherboards will experience so much inconsistency! Also amd usb controller has low polling rate, don't know currently value from head, but intel has 0.05ms. So some interrupts will be grouped together and than there will be a skip. In case of intel - this wouldn't matter and you can even edit that by editing some value in memory of some hardware register, even it is not easily done, as it different per piece of hardware - where it is. I see DPC latency being more of issue, a huge one!

Imagine all people, which will jump on hype train with shitty motherboards. I don't know what is avearge, but i would guess something between like 250-500. They will experience so much inconsistency LOL. Even it is still kinda high polling, dunno. Problem is, if you miss timer resolution window, than this will be brutal + if you have high dpc latency. Lets say in theory: you get update from mouse, after one window of time resolution occurs.... Correct me if i am wrong:

Time starting at 0, first timer resolution window 0.5ms. So lets say you get position from a mouse at time 0.51ms (counting dpc latency handling time by OS). So if you have like 250 DPC lat, it took 250us to handle that, so that would be 2 mouse positions back (not even considering polling of usb controller). So at 0.51ms it gets mouse position 2 polls back from time - instead of current one. Yet timer resolution window occured at 0.5ms. So it will render 3rd poll back from the past, which occurred before current one. That's latest mouse position OS handled and will be rendered in your game.

But than from 0.51ms, can occur only 3 polls, before next timer resolution window is at 1.00ms. So you get -1 poll from latest. And previously you would get -3 from latest. That's like 250us variation. Tho you have to remember timer resolution code updates to the cpu happen periodically at constant rate. Also usb controller from intel polls each 50us. Also no reason to think, these are synced. But mouse event, now i am talking about mechanically moving a mouse, it can occur in any period of time, before, or after each window of these two systems. So there will be variation. Even at 8k, variation will be significantly smaller in terms of hundreds of us.

Nevertheless, unless you have DPC lat. below 100us, you can't even utilize 8k polling rate. And you have to also count usb controller, which polls at 50us (for intel), so if you hit exactly at 51us, you will get additional 49us of delay - making this 149us, which is above 125us. So you need dpc. latency <= 76 and i would like to see that, to get motherboard like that! Amd has one 55us mobo b550 i think, but amd controller has low polling rate! So you may need to edit memory in registry of a usb controller, to even utilize this, which is risky and hard to find, if you are not expert, every cpu has it elsewhere. Not to mention it is already hard to pick mobo and you will be limited to couple choices and not everyone has money for that!

Hope you got it, because i am terrible at explaining...

HAHAHA people gave me sh1t, for saying that DPC latency is important and now they will get a mouse, they won't be able to run on their motherboards... They should make new interface for mouse honestly, usb3 has high latency and feels terrible. USB 2 is superior for gaming! But modern mobos don't have even usb 2 drivers anymore...

So better watch anandtech for low DPC latency motherboards :lol: :lol: :lol:

O_O
Posts: 19
Joined: 26 Oct 2020, 06:19

Re: I have the new Razer 8000 Hz prototype gaming mouse on my desk.

Post by O_O » 05 Nov 2020, 00:48

empleat wrote:
04 Nov 2020, 11:02
Also usb controller from intel polls each 50us.
I think you mean interrupt moderation as Intel polls to 125us. Polling can occur between USB Host and device, interrupts occur between USB Host and CPU. They are not dependent on timer resolution, for instance one can record 8k mouse data with 0.125ms intervals with MouseTester while using a timer resolution of 15.6ms.

I take it you mean EHCI and XHCI when referring to USB2 and USB3 as XHCI supports both USB2 and USB3.

Hope your RSI heals soon.

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

Re: I have the new Razer 8000 Hz prototype gaming mouse on my desk.

Post by empleat » 05 Nov 2020, 04:58

O_O wrote:
05 Nov 2020, 00:48
empleat wrote:
04 Nov 2020, 11:02
Also usb controller from intel polls each 50us.
I think you mean interrupt moderation as Intel polls to 125us. Polling can occur between USB Host and device, interrupts occur between USB Host and CPU. They are not dependent on timer resolution, for instance one can record 8k mouse data with 0.125ms intervals with MouseTester while using a timer resolution of 15.6ms.

I take it you mean EHCI and XHCI when referring to USB2 and USB3 as XHCI supports both USB2 and USB3.

Hope your RSI heals soon.
Yeah i heard someone said 50us, than someone 8k polling rate... In intel XHCI datasheet is 8k, not sure if it is maximum value tho, or only when IMODI register is set to value - 512! https://www.intel.com/content/dam/w...e ... b-xhci.pdf

Yeah so far as i know, usb controller polls mouse constantly and than sends interrupt to a cpu, which than schedules DPC. Yes they are not dependent on the timer resolution. (tho i heard, lower timer resolution interval can decrease DPC latency, don't know, it might indirectly, because it increases cpu usage, so cpu is able to process things faster and get back to DPC). That's not how i meant it. But than once DPC is handled by OS, code needs to be updated into cpu!!!!!!!!!! Try timer resolution on Windows 10, it can be set lowest to only 0.5ms interval. Even 0.5ms vs 1ms, you can notice huge reduction in input lag!!!!!!!! You can try it, keep disabled dynamic tick tho, cause mouse acceleration! So when your mouse position will render is still bottlenecked by the timer resolution, which is 0.5ms maximum value!

Also each of these systems are not synced and captured data in various intervals, for example: what if your DPC is handled by OS right after timer resolution interval occurs?!

Ty, tho i have it years by now, nerves heal extremely slow...

User avatar
AddictFPS
Posts: 314
Joined: 15 Jan 2020, 14:13

Re: I have the new Razer 8000 Hz prototype gaming mouse on my desk.

Post by AddictFPS » 05 Nov 2020, 15:03

Windows 7 also 0.5ms cap. Monitor frequency 512Hz cap. Windows 10 update should fix both.

If games in FullScreen mode can take almost all resources, i wonder if maybe can override OS 0.5 limit, to apply temporaly what game engine consider needed. Or maybe is just avoiding desktop composition layer. The fact is than ingame, i suppose in FullScreen, NDUS say the same 8K polling doing mouse circles, use less CPU load. Maybe is related, and OS cap only affect windowed games.

User avatar
Chief Blur Buster
Site Admin
Posts: 11653
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: I have the new Razer 8000 Hz prototype gaming mouse on my desk.

Post by Chief Blur Buster » 06 Nov 2020, 00:16

AddictFPS wrote:
05 Nov 2020, 15:03
Monitor frequency 512Hz cap. Windows 10 update should fix both.
Please be careful about parrotting this contextless informatoin. Not for a while; the feature is only for vendors at the moment for long-term incubation first before being available in retail Windows Update. It will take time for unlocked 1000 Hz to hit retail Windows build.
AddictFPS wrote:
05 Nov 2020, 15:03
Windows 7 also 0.5ms cap.

Please be careful about parrotting this contextless information. There are dozens of different timers/counters/etc of various kinds of precision. On some systems, event timers can occur at relatively microsecond-league precision, while other systems are limited by the 0.5. I have one system where the standard C# .NET event timer is sub-0.5ms, and on another system there's a 0.5ms granularity occuring. It's not a universal behavior.

Thank you. :D
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!

Locked