Placebo or a possibility? USB ports and mouse feel.

[ashrr]

For quite some time now I've had an issue where my mouse would feel great for a few days then the for the next couple of days it would feel kinda loose and icy. I have tried to fix this in the past by trying some common and not so common tweaks in bios and windows but nothing has ever fixed it permanently so kinda gave up and chalked it up to getting older and not playing as much.

Now I'm not so sure. A week ago I was playing a game, going through a mouse feels terrible patch, needed to straighten the mouse cable a bit so unplugged it for a few seconds and after plugging it in again it felt sharp and accurate. Finished my game wondering wtf was going on then I thought I would try unplugging/plugging my mouse in and it felt icy again. Unplugged it from the top intel usb ports marked for keyboard and mouse and into an asmedia port and it felt great and has been for about a week now.

I'm willing to accept that this is all placebo but if it isn't why or is it even possible for a mouse to feel different in different usb ports?

Some info and specs:
9900k @ 4.8GHz
16GB DDR4 @ 3600MHz CL16
GTX 1080 (Latest drives, low latency on ultra)
Z370 Fatal1ty K6
PG258Q @ 240Hz (No Gsync or Vsync)
Model O @ 1000Hz (Was Intel USB port now ASMedia)
G413 (Intel USB port)
5 Internal Drives (Intel SATA ports)
2 USB Drives (Intel USB ports)
Windows 10 1909 (Clean install about 2 months ago)

[Mr1991]

Yes it does feel different in different USB ports, but as you say a lot of that is from taking out/rearranging USB ports when windows is active, usb 2.0 is generally considered more responsive than 3, 3 feels like it has a slight “smoothing” effect to it, but yes just by taking it out and putting it back in the same port alters mouse movement, makes it feel “lighter”, Feels like you can reach your target quicker, but maybe harder to control, and what ports you use does play a slight role, but more so what you have connected around it, i use the back top 2.0 usb for mouse, but if I plug my keyboard in next to it, it dulls down my mouse movement, even after restarting, so I get best results by putting the keyboard away in the bottom usb 3, or in one of the front ports, you’ll hear some ppl say “don’t do this don’t do that”, but a lot of it is different system to system

[skylit]

You cant really generalize this because different chipsets and 3rd party controllers are used on a mobo to mobo basis.

The new RYZEN 3000 cpus generally run 2-4 Direct CPU line 3.1 gen1/2 ports to the back panel where the rest come from the chipset and or 3rd party controllers. Intel is currently running 100% through chipset and or 3rd party controllers depending on implementation of mobo.

USB 2.0 feeling better could be the result of more or less lag and an optimal path rather than the "default" choice. On Ryzen, rear panel 2.0 is most likely handled by a 3rd party controller where the native 4x 2.0 ports are used via on board connectors.

Just use whatever feels best is the tl;dr

[Sparky]

Could be placebo, could be real. Only way to find out which is true is to actually measure the performance on your specific system.

[ashrr]

@Mr1991
I wonder why this is the case tho, it would be nice to be able to actually test this. I've seen loads of threads on oc.net from roach talking about usb ports and maybe there is something to it but he's usually dismissed due to lack of evidence.

@Sparky
I've thought about getting a high speed camera before but didn't bother due to price and then just kinda accepting maybe it's just in my head. It also doesn't feel like straight up more input lag either so not sure how useful a high speed camera would be.

[Mr1991]

Yeah, as you say it’s not necessarily input lag, rather how it handles mouse movement, 2.0 feels more sensitive to small movements/accentuates jitter, like you’re on caffeine, the difference between a Razer mouse and a Zowie mouse for example, why something changes when you take out/re insert a mouse who knows, personally I don’t believe Windows is true untouched 1:1 even without enhanced pointer precision/acceleration fix, when you re-plug a mouse, it probably alters/bypasses this in a way, you can get a similar/view this effect by going into device manager and expanding human interface devices, when you take out and re-apply a mouse, this page refreshes/a lot of things get refreshed/re applied, simply just deleting one of these device listings has a similar effect as re-plugging your mouse, so my guess is the sorting system of all these devices get shuffled, which in turn maybe alters the priority/the way Windows reads it.

[TTT]

OT but why are you using Ultra LLM on that setup?

Surely you aren't GPU bound and getting 60-100FPS.

[Chief Blur Buster]

Yes, USB jitter is a big problem on some ports. During measuring, I've noticed crappy USB performance with latency surges to 4ms-timescales especially during hub contention. Often, the adjacent motherboard ports behave like a hub. That's a problem if you use the same ports for other devices, like a thumbdrive, external disk, VR headset, or other devices that compete for USB resources.

Some professional players make sure there's only one device connected (only mouse, or only keyboard) per hub -- and never share high-data-rate devices like Oculus Rift.

One extremetweaker trick is to purchase a StarTech 4 port USB3 card, and move all your USB devices to it, while keeping motherboard ports only for your mouse/keyboard. This helps guarantee ultra-low USB latency jitter for your favourite gaming mouse, with no interference from things like Oculus VR headsets, USB speakers, or external hard disks, etc.

I'm even wondering -- scientifically -- if USB jitter can potentially be one of the weak links in the Vicious Cycle Effect (say, strobed 4K) where the competing higher resolution, wider FOV, faster game motion, reduced motion blur, bigger displays, closer viewing distances, all amplify each other to make those new sub-millisecond errors even more human visible (e.g. microstutter is even more visible during 0.25ms MPRT than it is at 2.0ms MPRT). I bet this can be helpful if you overclock to 2000Hz+ or use the Cougar Mios X5 mouse, where USB jitter starts to become a noticeable error margin. At high motion speeds (4000 pixels/sec with a TestUFO-smooth high-DPI mouse), with ultralow MPRT, microstutter weak links are especially amplified.

More tests are needed.

[TTT]

Yes, USB jitter is a big problem on some ports. During measuring, I've noticed crappy USB performance with latency surges to 4ms-timescales especially during hub contention. Often, the adjacent motherboard ports behave like a hub. That's a problem if you use the same ports for other devices, like a thumbdrive, external disk, VR headset, or other devices that compete for USB resources.

Is there a way to measure latency on ports?

I have a Asus Tuf Z370 with 2 2.0s, 2 3.0, 2 3.1 with ASMedia driver and 2 3.0 on the front panel.

[Chief Blur Buster]

Is there a way to measure latency on ports?
I have a Asus Tuf Z370 with 2 2.0s, 2 3.0, 2 3.1 with ASMedia driver and 2 3.0 on the front panel.

You can use an Arduino. Sending a ping from PC to a custom Arduino program, and the Arduino program pings back.

Incidentially, some gaming mice use the same chips (ARM 32-bit) as a modern Arduino!

You can also research the gaming mouse ARM chip, then hunt for an Arduino with a similar ARM chip, then write a custom Arduino IDE app whose only purpose is to respond to USB pings, then you'll have some rather interesting USB-port analysis. It's ugly when an external hard disk is also running at the same time on the same USB hub.