Benchmarking latency across common wireless links for microcontrollers (opens in new tab)

We nerd sniped ourselves into testing the latencies of a whole bunch of wireless communication links and protocols for microcontrollers.

We’ll probably do a series of power consumption / range tests later on, let me know if there are any setups in particular that you’d be interested in seeing test cases for.

Raw data, firmware and post processing scripts are here on GitHub:

https://github.com/Scottapotamas/embedded-wireless-latency-e...

This is one of the most thorough and well laid out analysis of useful hobbyist and low level research options out there. I'll be referencing this for years, thank you very much. Clearly a lot of effort has been put into this.

Years ago I was working on a system where an Android device communicated with a web server written in Erlang, and our Android guy was going off about how Erlang was "really slow" and that made no sense to me. So we looked at the interaction times, and it was certainly true that something was slow. Instead of the Erlang server, we swapped out Apache serving a static page. That was slow too. It turned out he'd been benchmarking the crappy wifi connection.

Something of a tangent but I couldn't find anywhere a way to wirelessly extend usb ports. Allowing you to plug a mouse/keyboard/Yubikey to a computer in another room with minimum latency and without the need of any software or particular drivers.

If one wanted to do something like that one would probably start by reading that article I guess.

Thanks for this very thorough work.

That is great stuff, many thanks for it. While not in such depth, for a precision timing network covering multiple sensors our analysis also showed that the delays and variation inherent in most higher-level radio protocols are far too high for microsecond-level synchronisation.

For this reason, we use the nRF52 radio peripheral directly (i.e. barebones C) which means we know exactly when the packet was sent and - at the other end - when it was received. Details are in the Production Specification document, which also shows how long it takes for the transmit to spin up (turn on, transmit pre-amble and so on).

Great article!

Here's a perspective from a guitarist's viewpoint. Most wireless guitar systems boast latencies below 3ms, with the better ones being closer to 1ms.

Latency is important here because even small delays can be felt by the player. Since sound travels at ~1 ft per 1ms, a delay on the order of BLE is equivalent to playing with the amplifier ~25ft away from you, which is pretty bad.

It is surprising that the best system is actually nrf24, given how old it is (or maybe because of that ?). It also seems to have enough bandwidth to transfer a 24-bit/48kHz signal.

Thanks for sharing your research!

I’ve done some similar, but not as thorough, tests with 2.4ghz WiFi and 915MHz LoRa before. My goal is to sync time across multiple devices that go in and out of range of each other to play 60fps light animations and sound within half a frame of each other (8ms), with spare time for some computation.

I’ve been surprised at how bad some oscillators (or voltage regulators) can be and their effect on consistent latency.

I was having fun experimenting, but this will be really useful to get me back to actually implementing my project.

Outstanding blog post. NRF24 had been my fav for a long time, I planned switching to LoRa but I didn't know the latency is that high.

This is a great breakdown, although it doesn't go into any detail on the distance trade offs of each transmission type, which is an important part of the comparison. I look forward to when they go into this part of the analysis.

If we had more great reviews like this we'd see hardware improve faster! Thank you! Shine that light!

Amazing article and research. Most definitely bookmarked.

Super cool. I'd love to see 2.4GHz LoRA as well.