Mouse Click Latency

JacobKS · October 17, 2021

Hello,

Regarding the Razer Viper Ultimate and Razer Viper 8k hz click latency. Razer advertising the Viper 8k hz as 0.125ms click latency.

Meanwhile on this website the comparison of the 2 products are different to what was advertised. https://www.rtings.com/mouse/tools/compare/razer-viper-ultimate-vs-razer-viper-8khz/1639/22273?usage=8876&threshold=0.10

Under the Click Latency section it says for the Viper Ultimate when using wired is at 7ms and for the Viper 8k hz itself is at 6ms.

Question:

1. Why on that website the Viper 8k hz did not reach the advertised speed?
2. Which is correct?
3. Which of them is accurate?

Regards,

Jacob

WereCat · October 17, 2021

12 minutes ago, JacobKS said:

Hello,

Regarding the Razer Viper Ultimate and Razer Viper 8k hz click latency. Razer advertising the Viper 8k hz as 0.125ms click latency.

Meanwhile on this website the comparison of the 2 products are different to what was advertised. https://www.rtings.com/mouse/tools/compare/razer-viper-ultimate-vs-razer-viper-8khz/1639/22273?usage=8876&threshold=0.10

Under the Click Latency section it says for the Viper Ultimate when using wired is at 7ms and for the Viper 8k hz itself is at 6ms.

Question:

1. Why on that website the Viper 8k hz did not reach the advertised speed?
2. Which is correct?
3. Which of them is accurate?

Regards,

Jacob

Because of the testing methodology.

The Razer spec is the "in theory on paper you can achieve 0.125ms because 1000 (Hz polling rate of the USB controller) / 8000 (Hz polling rate of the mouse) = 0.125ms" but the true delay from the mouse click to what appears on the display depends on the signal travel time from the mouse to the PC, then from the PC to the display. You don't even know how they tested the mouse (if it's on the website I haven't read it) so depending on how much Hz your monitor has and how much FPS your game is able to run at, the difference in testing may be even 10ms between the results.

So that's why in theory the 1000Hz polling rate vs 8000Hz polling rate makes no actual difference because the improvement is at best 0.875ms so if you take into account the range of error during the actual use case, it can be as much as 5ms to 10ms just depending on what kind of PC you have, how much FPS you've got and how much Hz your monitor has.

TheChromaBristlenose · October 17, 2021

Rting's latency measurements are incredibly flawed, and entirely useless for basically anything. With one PC, they have the mouse click trigger a speaker, and by recording on a second PC and visualizing the output with a spectrogram, they compare the time from when the mouse "clicks" to when the speaker plays. This doesn't sound like too bad of an idea to begin with, but the audio pipeline is notoriously prone to latency fluctuations which causes the margin of error to be larger than what they're trying to measure in the first place. Also, with two PCs they've effectively introduced twice the number of variables they have to control when it comes to latency. That's not all either - on mice with optical switches (like the Viper lineup), the light shutter doesn't always close at the same instant the click is generated, which invalidates any audio-based comparison for them even if it was done properly.

Their measurements suck to the point that their database isn't even internally consistent - if you look through it, the Naga Pro in Bluetooth mode (notoriously slow wireless protocol) is shown to be faster than basically all wired mice, the Burst Pro/Core (which have identical PCB and firmware where clicks are concerned) have wildly different measured latencies, several wireless mice are measured as significantly slower when plugged in, etc.

Now to your question - the reason why Razer advertises such a low number is because they're only accounting for the isolated peripheral latency, which should at maximum be the polling interval (0.125ms in an 8000hz mouse) assuming that it's not being delayed by something else (usually debouncing). Razer doesn't advertise the end-to-end latency, because that number will vary enormously from system-to-system based on factors like monitor refresh rate, server tickrate, programs in the background, and a million other things. Their numbers they get on their test beds will not be the same as what most of their customers will experience.

It is possible to get a picture of what the Viper 8KHz's latency looks like beyond what Razer officially advertises, though. Using the bump test against the Ikari Optical (generally the baseline for such things), the 8KHz registers at under -1.0ms with a bit of fluctuation due to the polling rates of both. Nvidia Reflex indicates the Ikari is 1.5ms to real time, so Viper 8KHz is at least within the 0.5ms range (from click to PC registering). Razer's internal, unofficial testing with a USB analyzer (more accurate than bump test) shows 0.08ms to real time from when the poll hits, meaning the most realistic figure is just over 0.1ms from when you click to when the PC receives the input. All in all, we can quite confidently say that it is the fastest mouse on the current market by a good margin, even when taking Razer's marketing numbers with a good load of salt.