Video latency (lag) what is yours and what's your set up?

I get NO LAG using LG G4 with P3A and latest firmware upgrade and DJI GO
 
D
I get NO LAG using LG G4 with P3A and latest firmware upgrade and DJI GO
Yes you do.

To determine your lag, download a free stop-watch app to your cellphone that has 1/100 of a second accuracy.

Without props installed, indoors, turn on the Phantom and frame the phone (running the timer app) with the Phantom camera. No need to record video, just display the phone through the G4. While the timer is ticking away in the G4 screen, take a photo with another camera that frames the phone running the timer in the foreground AND the screen of the G4 in the background.

When you snap a photograph the G4 screen that is displaying the timer, AND in the foreground showing the phone screen w/timer, you can use those two numbers in the photo to calculate latency to the millisecond.

Bottom line, it's impossible to capture, compress, transmit, decompress and display on a screen without lag using h.264 in Lightbridge in Phantom 3 P/A. You will likely see about 150 to 250ms latency using Lightbridge 1.
 
Last edited:
  • Like
Reactions: Vertigo
D
Yes you do.

To determine your lag, download a free timer app to your cellphone that has 1/1000 of a second accuracy.

Without props installed, indoors, turn on the Phantom and frame the phone (running the timer app) with the Phantom camera. No need to record video, just display the phone through the G4. While the timer is ticking away in the G4 screen, take a photo with another camera that frames the phone running the timer in the background AND the screen of the G4 in the foreground.

When you snap a photograph the G4 screen that is displaying the timer, AND in the background showing the phone screen w/timer, you can use those two numbers in the photo to calculate latency to the millisecond.
I think what he means is no discernible lag. Mine's the same, obviously there will be some lag, but it's so insignificant that it I don't really notice it.
 
I think what he means is no discernible lag. Mine's the same, obviously there will be some lag, but it's so insignificant that it I don't really notice it.
Obviously might be different if I was racing through trees, but I tend to avoid them wherever possible!
 
I think the lag on the Nvidia Shield is about .5 or .3 seconds.. the same as you mention. I don't see this as being an issue at all. Jump to the 1 minute mark.

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.
Todd,
Just one question when you were conducting that test. Was the camera gimbal secured and fixed?

Because if it is not locked in place, you will notice when you turn your Phantom from side to side, the gimbal will slow down the turn of the camera, part of the 3-D stabilisation process. Thus the video will appear more lag when the gimbal is doing its job.

If your camera gimbal is secured, then your test is true.
 
The android app defaults to hw encoding if it's available. However, Litchi does give you a sw/hw option that you can choose yourself.
Thanks for the clarification on that!
 
I think what he means is no discernible lag. Mine's the same, obviously there will be some lag, but it's so insignificant that it I don't really notice it.
I agree, it's obvious to most of us using Lightbridge, we will always have video latency you can't fix, it's inherent to the technology. The OP was comparing to virtually no lag Immersion analog technology. So when someone says they have NO LAG (in capital letters) using Lightbridge, implying it's as fast as Immersion, they are obviously ignorant. Hence I tried to help out and explain he's mistaken to minimize confusion by others that do see the fraction of second lag (like the OP), and how it can be accurately measured for comparison.
 
Todd,
Just one question when you were conducting that test. Was the camera gimbal secured and fixed?

Because if it is not locked in place, you will notice when you turn your Phantom from side to side, the gimbal will slow down the turn of the camera, part of the 3-D stabilisation process. Thus the video will appear more lag when the gimbal is doing its job.

If your camera gimbal is secured, then your test is true.
My movement is far faster than the mechanics of the gimbal. Look at the initial movement of the Phantom and the video to get an idea of the delay. The camera would only move at the end of the Phantom movement.
 
Here's a more precise measurement of video lag between my P3A and an Android Nexus 7 tablet. Stop the playback and look at the delta between the time on the stopwatch app at the bottom and the display above after being sent via Lightbridge. I saw delays of 200 to 350 msec.

The first test is using DJI GO, and the second is with Litchi + HW decode.
To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.
To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.
 
  • Like
Reactions: CactusJackSlade
Wow, that's some interesting news, I'm trying to find if there are any tweaks that will help with this issue. The lag makes it a bit scary.

Anybody else using a Galaxy Tab S2?
Are you aware of the Zapper app for android? It basically stops most app's and sevices that's not needed and makes your tablet run faster and will have more memory and the cpu will not be bogged down. It made a hell of a difference on my nexus 7. Once your tablet has booted up wait about a minute or so until everything has loaded, then start the zapper app and hit select all and then hit kill apps.
 
These are not fpv racers. Don't you think the quad racers would use 720p video link also if there was no lag? .

Adammmmm - Finally the answer I was really looking for.... Everyone was saying "my tablet works great!.... very little lag" and I guess they have never used an FPV specific system so they probably had nothig to compare too :) ... and the HD downlink is fantastic, makes it hard to go back to my non-HD FPV goggles and quads...

This was what I was trying to figure out, like I said I'm coming from FPV background and I just wanted to know as you said "this is as good as it gets" as my other rigs are virtually instantaneous... and no I will not be racing this! LOL!

Also thank you (and everyone else) for the tips on running a "clean" tablet for maximum speed. I will calm down and enjoy my set up!

Cheers,

CJS
 
I agree, it's obvious to most of us using Lightbridge, we will always have video latency you can't fix, it's inherent to the technology. The OP was comparing to virtually no lag Immersion analog technology. So when someone says they have NO LAG (in capital letters) using Lightbridge, implying it's as fast as Immersion, they are obviously ignorant. Hence I tried to help out and explain he's mistaken to minimize confusion by others that do see the fraction of second lag (like the OP), and how it can be accurately measured for comparison.

John, Again, exactly the answer/info I was looking for - thanks! I had heard that Lightbridge had lag while I was heavy into all my FPV stuff so stayed away, I figured no lag on the Inspire or P3P was too good to be true, or as Adammmmm said all of us FPV racers would be using it ;-)
 
Are you aware of the Zapper app for android? It basically stops most app's and sevices that's not needed and makes your tablet run faster and will have more memory and the cpu will not be bogged down. It made a hell of a difference on my nexus 7. Once your tablet has booted up wait about a minute or so until everything has loaded, then start the zapper app and hit select all and then hit kill apps.

Yes, installed Zapper - I did a whole variety of things, one thing at a time, tested, re-tested the lag and virtually no difference. Even with Zapper I still got the same .3-.4 seconds latency :-( Oh well, I guess when someone say they have NO lag I will have to see it to believe it ;-)
 
As someone pointed out moving the 'copter to show lag is not the best way as the gimbal can cause delay as well.... here is how I test the lag and what I'm seeing on my Galaxy Tab S2:
To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.
 
CactusJackSlade:
I just did the stopwatch test on my 9.7" Tab S2 with my P3A. You can pause the video at any point to see the difference in both stopwatch. I am seeing approximately 0.25s lag.
To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.
 
  • Like
Reactions: CactusJackSlade
CactusJackSlade:
I just did the stopwatch test on my 9.7" Tab S2 with my P3A. You can pause the video at any point to see the difference in both stopwatch. I am seeing approximately 0.25s lag.
To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.
Perfect, nice example!.... and nice to put an actual number to it (mine was a BASIC test!) :rolleyes:
 
Here's a more precise measurement of video lag between my P3A and an Android Nexus 7 tablet. Stop the playback and look at the delta between the time on the stopwatch app at the bottom and the display above after being sent via Lightbridge. I saw delays of 200 to 350 msec.

The first test is using DJI GO, and the second is with Litchi + HW decode.
To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.
To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.
Excellent - thanks!
 

Members online

Forum statistics

Threads
143,066
Messages
1,467,355
Members
104,934
Latest member
jody.paugh@fullerandsons.