Have you ever wondered how much delay our VTX and VRX have? In this article we will test the latency of our 5.8Ghz analog FPV system.
In this test I used the same method which I used to test FPV camera latency test: FPV Camera Latency Testing
I tested a few different VTX (video transmitters) and VRX (video receivers) I have available. The same camera is used to ensure the results are consistent.
 |
I compiled the specifications of all 5.8GHz VTX for mini quad in this spreadsheet so you can compare them more closely. |
Testing Targets
VTX
VRX
- ImmersionRC Uno (one antenna)
- Dominator V2 + True-D V3.5 (diversity antenna)
I used the Runcam Eagle 2 as the FPV camera. VTX and VRX antennas are placed about 20cm away from each other. With the Dominator V2, I am getting the video signal from the “video out” port.
How much latency do VTX and VRX have?
| Testing Targets | Max (ms) | Min | Average |
| Eagle 2 | 33.8 | 15.2 | 24.6 |
| VTX03 + Uno + Eagle2 | 35.0 | 15.5 | 25.5 |
| Hawkeye + Uno + Eagle2 | 35.3 | 15.6 | 25.6 |
| VTX03 + True-D + Eagle2 | 34.6 | 15.3 | 25.2 |
| Hawkeye + True-D + Eagle2 | 34.8 | 15.2 | 25.3 |
The result shows that my VTX and VRX have less than 1ms delay in average, which is quite insignificant compared to the 20ms-30ms delay of most popular FPV cameras out there today.
It’s interesting to note that the the different VRX’s made a more noticeable difference to the delay than the VTX’s. But the difference is within 0.5ms which is nearly impossible to tell in real life.
Image processing is what cause delay in our FPV system. I think our result is reasonable since there is not much processing going on in these devices I tested. However it would be interesting to see if products like ClearView would have any impact to the latency, because they do much more processing on the signal which could introduce delay to the system.
Questions and comments
Would the distance between VTX and VRX distance change the result?
I don’t think placing VTX further away from VRX (or closer) would make much of a difference to the latency, since electromagnetic signals travels as fast as light. Sometimes we see some form of “delay” when signal is weak, just because of the dropping frames.
Do different VTX / VRX have different latency?
I think so, but I don’t think the difference is big enough to be concerned about.
Does OSD introduce latency to FPV video?
Not that I noticed in my tests, I tried it with and without OSD (Betaflight OSD on the FC) and there wasn’t a noticeable difference. (edit: MAX7456-based OSDs just draw over your video feed, that can’t cause any delay as original feed remains intact)
Have you done an end to end test of a FPV system that has image compression in then pipeline? e.g.- Sensor-compress-transit-receive-decompress-display. Have pro drone racing requirements at 0.020 seconds (“Racing Drones” Air & Space article). Appears this is being met by sending uncompressed video directly over a broadband connection. Is this being met similarly in consumer drones?
So the video relay will only add approx 1ms? Did I get it right?
Hello Oscar!
Thank you for your awesome work!
Did you do a comparison of the Latency of FPV goggles or monitors also? Or do you know where to find such a comparison? Would be interesting to see how much latency a Fatshark or skyzone or Aomway Goggle actually adds to the system.
Keep up your fantastic work!
Hey Achim
that’s a good suggestion, but the cost of doing that comparison would be much greater than that of FPV cameras. And the testing requirement might be different from what we have with cameras. I might do that in the future, not right now though.