I have a little bit of a back log of tests that I’ve gotten done, but just decided to work on it to get them done more quickly….

I had quite a few people ask me about crossfire latency after doing 2.4ghz frsky video. I finally got a crossfire setup and had some time to run my latency tests.

For a full description of the test, check out my previous test with frsky. https://youtu.be/hCX2Qi51ZtU

Short descriptions is I used a signal generator to send a throttle pulse every 1-1.5s. Pulse went over the air to be received by the RX and passed on to Betaflight.  Betaflight then increased all 4 motors. I have a logic analyzer on the motor line to record when the motors increased from idle by decode dshot commands.

Crossfire Latency Test
https://docs.google.com/spreadsheets/d/1LPpkwxKVqz1JMhiC_-FVhWYFZbWIHVbVZDnu8YpqSvI/edit?usp=sharing

Frsky Latency Test
https://docs.google.com/spreadsheets/d/1aTRuUzEIAqn2h_WNNHb7oMXqKZB5olL95rLlrjfThHw/edit?usp=sharing

End to End Latency

Opentx Crossfire 150Hz 9-21ms

Opentx Crossfire 50Hz 17-41ms

Ersky9x Crossfire 150Hz 9-19ms

DeviationTx Crossfire (testbuild) 8-26ms

Frsky 2.4GHz
Opentx FrskyX X4R Sbus 23-41ms

Opentx FrskyX X4R Fport 18-37ms

Ersky9x FrskyX X4R Fport 13-24ms

Ersky9x FrskyX SPI RX 11-22ms

DeviationTx FrskyX (Custom unreleased Build) 6-17ms

Ersky9x needs one of the recent builds with “double rates”
http://openrcforums.com/forum/viewtopic.php?f=7&t=4676#p65894

Crossfire is faster for most people moving from Opentx frskyX.  Ersky9x on a Taranis only a bit slower then Crossfire.  (Unreleased so far, but a slightly modified deviationTx is actually faster then crossfire)