evaluating xhci usb controller performance

[Hyote]

z690 tomahawk 12700k
Renesas uPD720202 is way better than intels xhci, using microsoft default driver
less usb coescalling? feels like rweverything imod

StarTech.com 4 Port USB 3.0 PCIe UASP - SATA / LP4 Power (PEXUSB3S44V)

I lost the screenshots but a guy on Discord tested these and concluded the default Intel controller to be superior to other USB controllers with ASMedia winning in a few tests that are not relevant in realtime use.

[kyube]

I lost the screenshots but a guy on Discord tested these and concluded the default Intel controller to be superior to other USB controllers with ASMedia winning in a few tests that are not relevant in realtime use.

I assume you're referring to https://twitter.com/BEYONDPERF_LLG/stat ... 04789956#m

The sample size (n=1) is too small.
The tester didn't provide repeatable steps to replicate his testing environment and thus allow others to contribute with their own hardware & OS to make up a larger sample size and give concrete conclusions.
His HW choices for the MXA tests are also not mentioned.
I do value his data, however. It's definitely something interesting to explore, as DPC/ISR driver performance is one of the most important parts in the journey of minimising CPU overhead from drivers (decreasing jitter).

I'm not sure what you refer to as “in realtime use”, as everyone seems to have their own definition of it

[Hyote]

I lost the screenshots but a guy on Discord tested these and concluded the default Intel controller to be superior to other USB controllers with ASMedia winning in a few tests that are not relevant in realtime use.

I assume you're referring to https://twitter.com/BEYONDPERF_LLG/stat ... 04789956#m

The sample size (n=1) is too small.
The tester didn't provide repeatable steps to replicate his testing environment and thus allow others to contribute with their own hardware & OS to make up a larger sample size and give concrete conclusions.
His HW choices for the MXA tests are also not mentioned.
I do value his data, however. It's definitely something interesting to explore, as DPC/ISR driver performance is one of the most parts in the journey of minimising CPU overhead from drivers (decreasing jitter).

I'm not sure what you refer to as “in realtime use”, as everyone seems to have their own definition of it

Nothing technical, just general computer use and gaming.

[1000WATT]

By the way, USB controllers can have their firmware updated.

z.jpg (382.27 KiB) Viewed 3666 times

[kyube]

By the way, USB controllers can have their firmware updated.

Yes, but not all PCIe AIC's have the ability to be updated using that particular application software.
They can brick the card

Off-topic: I've recently acquired 2 more AICs with these specific USB XHCI host controllers: Etron EJ168 & TUSB7340
I'll likely also acquire a TB3 (JHL7540) or TB4 (JHL8540) controller AIC.

I'd say that I have every 3rd party vendor USB 3.0 XHCI host controller available to the market for the past decade
I might skip USB 2.0 AICs due to the inability to run +1kHz on them without disabling IMOD.

My plan is to expand the testing data extensively on my opening post, using better ways to visualize the data.
No ETA.

[User_Friendly]

Hi all, I just want to add my input on something quite interesting incase it's gone under the radar for any of you guys.

M.2 to USB adaptor cards exist, allowing you to bypass the PCH on Intel systems and have USB input direct to the CPU via your PCIE M.2 channeled through the CPU.

I picked one up out of curiosity and it has been the single most consequential Intel platform improvement I've come across in all my years heavily invested in input latency/ input feel in windows.

The model I am using is: PH69 M.2 MKEY PCIe. With a Renesas UPD720201 controller, but there is another model available with an Asmedia ASM3142, I'll have one of those available to test shortly.

Also, taking XHCI USB controllers off of MSI mode and running line based provides a significantly better aim feel/ granularity for me. I don't know the mechanisms behind that, but anecdotally it's significant for me.

Additionally, I'd just like to shout out Hyyote, and all the others in this community. thankyou so much for your efforts, this endeavour of latency optimisation in windows is a punishing pursuit, and I really appreciate those doing so much work, and presenting their findings.

Thanks