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.
diakou
Posts: 83
Joined: 09 Aug 2020, 11:28

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

Post by diakou » 01 Oct 2020, 11:29

Chief Blur Buster wrote:
01 Oct 2020, 11:13
diakou wrote:
01 Oct 2020, 11:10
Kaldaien(Special K creator) back in 2018 (I assume he now has an even better understanding or different stance on it, made one of the first posts attempting to explain high polling rate dilemmas)
But it's now 2020, and Razer is working with Microsoft (they confirmed to me).
Indeed! I have been touting this for 2020 as I had been noticing a lot of low latency optimizations being in testing phases of 2019, it all got me curious as I've been on the latency warpath for a while now. Nvidia's LLM, AMD's equivalent. HAGS from Microsoft and other OS changes and improvements to their game mode. Their request in 2018 asking game devs to implement correct flip models for their games. Then we got Hall Effect keyboards commercialized to the mainstream through Steelseries and marketed as "low latency" from things such as the PCB, the keys mechanisms, setting actuation for raw human travel time latency improvement. The announcements of 360hz monitors in early 2020, NVIDIA's latency research/study in late 2019 to then be followed up with NVIDIA Reflex and one of the first massive-scale latency marketing operations.

There's many, many more things I have noticed that are leading up to this point, but it's making me very happy. 8k mice being introduced by non-mainstream companies pushed forward a lot of discussion about higher polling rates for mice and now we have something like Razer catch wind of it and attempt implementations. The good part about a lot of this, if something else in the system chain does not support X or Y, waiting for the support might take forever comparatively to innovators just pushing it and trying weird hacky methods to making things work. Now as you said, Razer are working with Microsoft to make this work smoother at an OS level.

I think with my limited understanding, the latency warpath will be going on by mainstream / big brands / companies for entire 2021-2022 and then from there I'm unsure what will happen. But with how much resources that seem to have been used in 2019/2020, I am simply estimating a 2 year timespan to justify the resources used. Excited to see some of the results from the mice later, thank you for sharing (even if that is probably marketing :mrgreen: )

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

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

Post by NDUS » 01 Oct 2020, 11:32

Chief Blur Buster wrote:
01 Oct 2020, 10:20

If any of you want to tell me additional tests to try with the mouse for this week's article, let me know!
A misconception I see all the time is that 8khz polling will consume vast amounts of CPU cycles

You can test this by changing the thread USB polling is performed on to a known thread using this tool: https://download.microsoft.com/download ... y_tool.msi - it must be ran as administrator, ignore any errors

You can also verify which thread polling is assigned to by using LatencyMon. The "CPUs" tab will show a list of interrupts on each core. Whenever you move the mouse you'll get tons of DPC counts on whichever core polling is currently assigned to. Be warned that unless you use the policy tool, the core responsible for USB polling can change when playing a game.

From here, forbid all programs from using the core responsible for USB polling using this elevated-permissions Powershell command:
$instances = Get-Process
foreach ($i in $instances) { $i.ProcessorAffinity=127 }
"127" will forbid all programs from running on core #8

You now have a core which is known to be only responsible for USB polling! You can observe its activity to see how much CPU a mouse will consume. I got worse results in the Windows desktop (~20% of a 9700K core w/ my 8khz M1K) than I did while in a video game (~5% of a 9700K core,) oddly

--

Also, re: software-side support for 8khz input
Here is a great post from an engine developer on Overwatch which details how he added support for high polling rates:
https://us.forums.blizzard.com/en/overw ... ion/422094


Here is a lua script you can use with Logitech's macro function to test whether a game supports high polling rates or not:
function OnEvent(event, arg)
if (event =="MOUSE_BUTTON_PRESSED" and arg == 4) then
MoveMouseRelative(126,0)
PressAndReleaseMouseButton(1)
MoveMouseRelative(126,0)
end
if (event =="MOUSE_BUTTON_PRESSED" and arg == 5) then
MoveMouseRelative(-126,0)
end
end
See https://www.youtube.com/watch?v=66C-FAy13DQ for how to add lua macros in Logitech G-hub

When you press mouse4, the gun should fire in the middle of the movement, on a non-rendered frame. If it buffers the click to the end of the movement, the game does not support polling rates higher than the rendered framerate. See https://streamable.com/ibif7a

User avatar
Chief Blur Buster
Site Admin
Posts: 11647
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 » 01 Oct 2020, 11:45

Excellent information, NDUS, about forcing poll Hz to a specific CPU core. For most games, I can sacrifice a CPU core to improving poll Hz reliability. Thank you.

Game developers and engine developers will probably get these prototype mouse before it hits the market, so engine developers can optimize popular engines and games (Unreal/Unity, and Fortnite/CS:GO/etc).
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
Razer_TheFiend
Manufacturer
Posts: 32
Joined: 01 Oct 2020, 11:51

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

Post by Razer_TheFiend » 01 Oct 2020, 12:00

We're actually looking at a Synapse software feature that would allow polling to be allocated to a single CPU core with "high" priority to minimize the impact, given how games are GPU bottlenecked anyways. No promises yet though, still an exploration.

ericl
Posts: 126
Joined: 09 Jan 2014, 14:50

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

Post by ericl » 01 Oct 2020, 12:48

That's actually a great idea (to make a core responsible for the polling). With modern CPUs having 8/16 cores, we have more than we need at the moment!

User avatar
speancer
Posts: 241
Joined: 03 May 2020, 04:26
Location: EU

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

Post by speancer » 01 Oct 2020, 13:00

When would you say we could expect 8000 Hz polling rate mice to hit the market?

Also, is 8000 Hz wireless achievable or there are currently technical obstacles that would prevent such an implementation? I got so used to high-end wireless mice that I don't wanna go back.
Main display (TV/PC monitor): LG 42C21LA (4K 120 Hz OLED / WBE panel)
Tested displays: ASUS VG259QM/VG279QM [favourite LCD FPS display] (280 Hz IPS) • Zowie XL2546K/XL2540K/XL2546 (240 Hz TN DyAc) • Dell S3222DGM [favourite LCD display for the best blacks, contrast and panel uniformity] (165 Hz VA) • Dell Alienware AW2521HFLA (240 Hz IPS) • HP Omen X 25f (240 Hz TN) • MSI MAG251RX (240 Hz IPS) • Gigabyte M27Q (170 Hz IPS) • Acer Predator XB273X (240 Hz IPS G-SYNC) • Acer Predator XB271HU (165 Hz IPS G-SYNC) • Acer Nitro XV272UKV (170 Hz IPS) • Acer Nitro XV252QF (390 Hz IPS) • LG 27GN800 (144 Hz IPS) • LG 27GL850 (144 Hz nanoIPS) • LG 27GP850 (180 Hz nanoIPS) • Samsung Odyssey G7 (240 Hz VA)

OS: Windows 11 Pro GPU: Palit GeForce RTX 4090 GameRock OC CPU: AMD Ryzen 7 7800X3D + be quiet! Dark Rock Pro 4 + Arctic MX-6 RAM: 32GB (2x16GB dual channel) DDR5 Kingston Fury Beast Black 6000 MHz CL30 (fully optimized primary and secondary timings by Buildzoid for SK Hynix die on AM5 platform) PSU: Corsair RM1200x SHIFT 1200W (ATX 3.0, PCIe 5.0 12VHPWR 600W) SSD1: Kingston KC3000 1TB NVMe PCIe 4.0 x4 SSD2: Corsair Force MP510 960GB PCIe 3.0 x4 MB: ASUS ROG STRIX X670E-A GAMING WIFI (GPU PCIe 5.0 x16, NVMe PCIe 5.0 x4) CASE: be quiet! Silent Base 802 Window White CASE FANS: be quiet! Silent Wings 4 140mm PWM (3x front, 1x rear, 1x top rear, positive pressure) MOUSE: Logitech G PRO X Superlight (white) Lightspeed wireless MOUSEPAD: ARTISAN FX HIEN (wine red, soft, XL) KEYBOARD: Logitech G915 TKL (white, GL Tactile) Lightspeed wireless HEADPHONES: Sennheiser Momentum 4 Wireless (white) 24-bit 96 KHz + Sennheiser BTD600 Bluetooth 5.2 aptX Adaptive CHAIR: Herman Miller Aeron (graphite, fully loaded, size C)

deama
Posts: 368
Joined: 07 Aug 2019, 12:00

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

Post by deama » 01 Oct 2020, 13:11

A test wherein you try to do flick shots would be nice, comparing 1000hz to a 8000hz. Both on an fps and rts (or desktop) would be nice.
Flick shots would be the ones where you quickly swipe the mouse and try and hit a target, probably should keep the target static.
Also, maybe do some mouse accuracy tests where you just click dots that come up on the screen and just compare the results and the general feel of it?

User avatar
Razer_TheFiend
Manufacturer
Posts: 32
Joined: 01 Oct 2020, 11:51

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

Post by Razer_TheFiend » 01 Oct 2020, 13:12

speancer wrote:
01 Oct 2020, 13:00
When would you say we could expect 8000 Hz polling rate mice to hit the market?

Also, is 8000 Hz wireless achievable or there are currently technical obstacles that would prevent such implementation? I got so used to wireless mice that I don't wanna go back.
Hard to say... The hardware is near-enough ready. We already know of some OS-level, controller-level and game-engine specific issues that we're already working with their respective owners to resolve. We don't want to release a product which has a very narrow operating window. An average customer will just perceive it as defective, since they don't care about the underlying root-cause of the issues. Once we resolve those issues, have a software developed and QA'd, we can go into production.

RE wireless : Not really feasible at the moment. Energy/Battery life is obviously one concern, and in addition, the wireless transaction time for our proprietary RF takes well over 125us (even though we're the fastest of anyone out there). Then you need some time for redundancy/recovery of lost-packets. > 1000Hz is definitely possible, but 4000 might be a bridge too far at the moment.

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

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

Post by NDUS » 01 Oct 2020, 13:50

deama wrote:
01 Oct 2020, 13:11
A test wherein you try to do flick shots would be nice, comparing 1000hz to a 8000hz. Both on an fps and rts (or desktop) would be nice.
Re: taking video of 8000hz polling, it's an ambitious task but can be done.

You can actually recompile OBS to let it record framerates of 10,000+ just by replacing "120" with "999999" in this file: https://github.com/obsproject/obs-studi ... ettings.ui - or if you don't wanna recompile you can use a memory editor (ie. cheat engine) to change the FPS integer

I've tested it up to 2400fps @ 360p and it worked fine, with the NVENC engine of my 1070 at 93% utilization. You need to use a resolution small enough (for a given framerate) that the NVENC encoder doesn't get overloaded (you can check its utilization with HWInfo64, the relevant value is called "video engine") - or you can use lossless / non-encoded video, but better have a lot of storage for that.

With this process, the frames that are rendered on the GPU get encoded to a smaller file weight & exported directly into a video. So if your GPU is rendering 8000fps, the video can be 8000fps (at a small resolution, of course - if you're using NVENC)

You also need a program that can output 8000+ FPS obviously. Some kind of barebones test program would do it. With this setup, you could directly record 1000hz and even 8000hz polling intervals in a 3D environment.

Any program that uses WM_MOUSEMOVE (the Windows cursor movement API - by example, Starcraft uses it) will fully support 8000hz polling. It's just a matter of finding or making a program that outputs 8000+ FPS and uses WM_MOUSEMOVE

deama
Posts: 368
Joined: 07 Aug 2019, 12:00

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

Post by deama » 01 Oct 2020, 14:12

NDUS wrote:
01 Oct 2020, 13:50
deama wrote:
01 Oct 2020, 13:11
A test wherein you try to do flick shots would be nice, comparing 1000hz to a 8000hz. Both on an fps and rts (or desktop) would be nice.
Re: taking video of 8000hz polling, it's an ambitious task but can be done.

You can actually recompile OBS to let it record framerates of 10,000+ just by replacing "120" with "999999" in this file: https://github.com/obsproject/obs-studi ... ettings.ui - or if you don't wanna recompile you can use a memory editor (ie. cheat engine) to change the FPS integer

I've tested it up to 2400fps @ 360p and it worked fine, with the NVENC engine of my 1070 at 93% utilization. You need to use a resolution small enough (for a given framerate) that the NVENC encoder doesn't get overloaded (you can check its utilization with HWInfo64, the relevant value is called "video engine") - or you can use lossless / non-encoded video, but better have a lot of storage for that.

With this process, the frames that are rendered on the GPU get encoded to a smaller file weight & exported directly into a video. So if your GPU is rendering 8000fps, the video can be 8000fps (at a small resolution, of course - if you're using NVENC)

You also need a program that can output 8000+ FPS obviously. Some kind of barebones test program would do it. With this setup, you could directly record 1000hz and even 8000hz polling intervals in a 3D environment.

Any program that uses WM_MOUSEMOVE (the Windows cursor movement API - by example, Starcraft uses it) will fully support 8000hz polling. It's just a matter of finding or making a program that outputs 8000+ FPS and uses WM_MOUSEMOVE
I was actually thinking in more normal terms rather than extreme, was thinking of if the chief could verify, or try to at least, what this guy did in the video and just explain whether it was easier for him at 8000hz or not:
https://www.youtube.com/watch?v=mcsbDDyfLLU
from the video, it looks like it was much easier for him even at not being able to get the fps that high (says 240-ish fps on top left screen).

Locked