Mouse polling rate

If you have invested in a high resolution mouse, adjusting the USB polling rate is a common trick to utilize the added precision it brings. The polling rate (or report rate) determines how often the mouse sends information to your computer.

Polling rate and polling interval

The polling rate of a device is measured in Hertz (Hz) and is determined by the polling interval. The polling interval is measured in milliseconds (ms) and equates to lag time.

The default polling interval is 10ms. However, USB controllers round the interval down to the nearest power of two. Thus, an interval setting of 10ms will actually use 8ms, 7ms will use 4ms, etc.

The following table shows the relation between polling rate Hertz and the corresponding interval milliseconds (rate = 1000 / interval).

Hz 1000500250125
ms 1248

If the polling rate is 125 Hz, the mouse position will be updated every 8 milliseconds. In situations where lag is critical — for example games — some users decrease the interval to as little as possible. However, this puts more load on the CPU, so care should be taken when adjusting this value.

Display polling rate

The evhz tool can display the actual mouse refresh rate.

You can install it from evhz-gitAUR and execute as root:

# evhz

Now move the mouse continuously in large circles until the displayed Average stabilizes then press Ctrl+c to exit.

If the Latest value does not stabilize and switches between two values, then the attempted polling rate is faster than the device is capable of; see #USB device speed.

Alternatively, Windows tools such as DirectX mouserate checker can be run using Wine. Or use a website based checker like the one provided by CPS-Check.

Display polling interval

Note: This only shows the polling interval requested by the device and not the actual interval being used. See BBS.

Device information including polling interval can be found in debugfs if it is mounted and you have root access.

First, find the vendor and product IDs of your device with:

Then run the following as root with those IDs to display the debug information for that device:

The Ivl is the polling interval; this device has requested 10ms (and actually reports every 8ms as explained in #Polling rate and polling interval). The is the device speed explained in #USB device speed. For information about the other fields, see the kernel documentation.

If debugfs or root access are not available, the polling interval can be shown with:

USB device speed

USB devices are designed to operate at a certain bitrate. Many pointing devices are "Low Speed" 1.5Mbit/s devices. The speed of a device can be shown as explained in #Display polling interval.

"Low Speed" devices may not be capable of polling at intervals less than 8ms.

All USB hubs should be capable of at least "Full Speed" 12Mbit/s. The speed of the hub that the device is attached to can be shown with the following command with the same Bus=xx as the device:

The Ivl of the hub is independent of the device and does not affect the polling rate of the device.

Set polling interval

To configure the polling rate, use the option of the kernel module. The default value is 0 which means the module uses the interval requested by the device(s).

The current value of the option can be verified with:

To change the configuration, create the following file:

This example requests a polling rate of 250Hz. Similarly, you may use jspoll or kbpoll to change the polling rate of gamepads/joysticks or keyboards.

To change the polling interval without rebooting

# modprobe -r usbhid && modprobe usbhid

You may have to unplug the mouse and plug it back in for the change to take effect.

Known issues

Polling at half of requested rate

There is a kernel bug that for certain configurations prevents devices from reaching 1000 Hz (1ms) polling rate. See the BBS and Bug.

A work-around that may help is to connect the device to a port using a different driver.

Polling rate not changing

The USB 3 driver xhci-hcd may be ignoring the setting. See the linux-usb mailing list message and Bug.

The xhci-hcd module should respect the interval requested by the device, so check the documentation for the device for a hardware or firmware setting.

A work-around that may help is to connect the device to a port using a different driver.

Another work-around is to disable xHCI. There might be a BIOS setting for this or you can do so by blacklisting the xhci-hcd module. However, either way will cause any USB 3 ports to act as USB 2 as the kernel will use the module instead.

As an alternative, one can install a kernel module wmo_oc-dkmsAUR (or build it manually from wmo_oc-kmod), which is a fork of gcadapter-oc-kmod for wmo1.1a (MS Wheel Mouse Optical 1.1a). The identifies a mouse with its vendor and product ID (VID:PID) values to overclock it even with XHCI hubs, without loosing USB 3 functionality. Changing the polling rate of any USB mouse other than the default wmo1.1a requires modifying the default VID:PID values (045e:0040) found in the module's source file in and finally rebuilding (removing and installing) the kernel module using DKMS.

Polling rate resulting in lag with wine

Having a mouse with a high poll rate and using wine can result in game lag; see here. It would seem that the user who created that bug had the problem solved by updating the bios. The bug, however, is not resolved for other users.

It is not possible to change the polling rate of the mouse using the methods within this wiki (the "usbhid" method) if your computer only has a USB3 xHCI Controller.

Unfortunately, there is currently no fix for users with a combination of a mouse with a high polling rate and only a USB3 xHCI Controller.

A workaround is to use a mouse with a lower polling rate.

gollark: I don't think I could read text easily if I had a denser-than-FHD panel (without accursed scaling things).
gollark: Or speakers or trackpad or whatever.
gollark: I can't see what actual *advantage* I would get out of a slightly nicer screen.
gollark: It's shiny and cool for some time, then you get used to it and can no longer put up with worse ones.
gollark: I avoid getting particularly "nice" ones because they cost more for less performance and also something something hedonic adaptation.

See also

This article is issued from Archlinux. The text is licensed under Creative Commons - Attribution - Sharealike. Additional terms may apply for the media files.