Viofo A129 Duo duplicating/jumping frames on rear camera/Firmware Bug Discovered

Guessing this is affecting everyone, but you were first to notice... Has anyone flashed back to version 1.3 to see if problem still exists?
I will test it but not until weekend.
 
Anxious to see if this issue exists in Firmware Version 1.3, software bug, or if this is an issue caused by hardware incapable of handling two streams at once.
 
Anxious to see if this issue exists in Firmware Version 1.3, software bug, or if this is an issue caused by hardware incapable of handling two streams at once.
If the hardware was not capable, it would drop frames, not duplicate them. And the fact that it duplicates the frames in the same way even at almost double bitrates (see 22/22 Mbps MODs) suggest that the hardware is not an issue.
 
If the hardware was not capable, it would drop frames, not duplicate them. And the fact that it duplicates the frames in the same way even at almost double bitrates (see 22/22 Mbps MODs) suggest that the hardware is not an issue.

The mods use the stock firmware so whatever issue is with the Camera or Firmware is only replicated with the Modded Firmware. I think you make sense about dropping frames, though I can't really fathom what would cause frame duplication either.
 
The mods use the stock firmware so whatever issue is with the Camera or Firmware is only replicated with the Modded Firmware. I think you make sense about dropping frames, though I can't really fathom what would cause frame duplication either.
Frame duplication is clearly a software bug. I don't know how to put it clearly, but my guess is that for some reason the front and rear videos get slightly out of sync at some point and they duplicate a frame every now an then to "wait" for the front camera and get back in sync. I guess that is intentionally used as a workaround to avoid rewriting a bigger portion of the code...
Having said that, I can live with that. I would be more concerned if there were dropped frames. The purpose of dash cameras is to provide evidence in case of an accident or as footage for road rage, vandalism, etc. In my opinion, a camera that duplicates some frames but doesn't drop any, does this job. Of course, it would be nice if @viofo will find a way to fix this minor inconvenience.
 
Last edited:
Frame duplication is clearly a software bug. I don't know how to put it clearly, but my guess is that for some reason the front and rear videos get slightly out of sync at some point and they duplicate a frame every now an then to "wait" for the front camera and get back in sync. I guess that is intentionally used as a workaround to avoid rewriting a bigger portion of the code...
Having said that, I can live with that. I would be more concerned if there were dropped frames. The purpose of dash cameras is to provide evidence in case of an accident or as footage for road rage, vandalism, etc. In my opinion, a camera that duplicates some frames but doesn't drop any, does this job. Of course, it would be nice if @viofo will find a way to fix this minor inconvenience.

Look on the front camera video between 1m:30s-1m:44s on my post #11.This kind of frame drop is not acceptable on a very important moment in the footage..Yes,the rear one only duplicates and will not affect the proof when needed,but the front one drops frames randomly..
 
Frame duplication is clearly a software bug. I don't know how to put it clearly, but my guess is that for some reason the front and rear videos get slightly out of sync at some point and they duplicate a frame every now an then to "wait" for the front camera and get back in sync. I guess that is intentionally used as a workaround to avoid rewriting a bigger portion of the code...
Having said that, I can live with that. I would be more concerned if there were dropped frames. The purpose of dash cameras is to provide evidence in case of an accident or as footage for road rage, vandalism, etc. In my opinion, a camera that duplicates some frames but doesn't drop any, does this job. Of course, it would be nice if @viofo will find a way to fix this minor inconvenience.

Then again, if the camera is pausing due to an under powered processor, I wonder if the lag then causes the camera to write frames twice. Which would be possible, because RAM temporarily stores an action. So maybe the RAM has the video buffered into memory, but when the processor lags, the ram causes a duplication of frames, at the point where the processor lagged and then begins writing to the card again.
 
Then again, if the camera is pausing due to an under powered processor, I wonder if the lag then causes the camera to write frames twice. Which would be possible, because RAM temporarily stores an action. So maybe the RAM has the video buffered into memory, but when the processor lags, the ram causes a duplication of frames, at the point where the processor lagged and then begins writing to the card again.
In that case you would have a missing frame somewhere, whereas it appears you just have an extra copy of one frame, with the shutter timing consistent between all the others, which suggests that the rear camera is collecting frames reliably but with a clock that is running slightly slow, so occasionally it doesn't deliver the next frame in time to keep up with the front.
 
In that case you would have a missing frame somewhere, whereas it appears you just have an extra copy of one frame, with the shutter timing consistent between all the others, which suggests that the rear camera is collecting frames reliably but with a clock that is running slightly slow, so occasionally it doesn't deliver the next frame in time to keep up with the front.

Gabilondo is saying the front camera drops frames. I haven't verified this myself, but he was right about frame duplication. And I would still argue that the video could be buffered into Ram just long enough for processor to pause for a split second and ram to hold the video. So then when the video is written again, it writes from the last spot where the ram hold the video and processor paused.

Especially now with the front camera dropping frames and rear camera duplicating. It's like the one is playing catch up and the other halts during process.

Look on the front camera video between 1m:30s-1m:44s on my post #11.This kind of frame drop is not acceptable on a very important moment in the footage..Yes,the rear one only duplicates and will not affect the proof when needed,but the front one drops frames randomly..
 
I saw one year old footage of somebody's A129Duo on YT,and the rear camera didn't have the frames problem,I truly hope this is a software bug that we have..

Viofo A129DUO IR have the same problem as ours,I can clearly see the frame duplicate every 3 seconds in this guy's footage.You have to look outside of the car..

 
Last edited:
Just waiting for someone to try out Firmware 1.3 and see if the Rear Duplicates and Front Drops....Don't have a spare card to run test and not wanting to format the one I'm using. Soon as someone tries out 1.3 we'll know more.
 
Anyone tested firmware version 1.3 to see if the problem still existed?
 
As I promised,I have downgraded today at v1.3 and the problem is still there..Front camera seems to be ok,but this one is failing randomly some clips yes,some clips no,with other firmware versions too.Rear camera have exactly the same issue like before,duplicating a frame,every 93 frames,same as with other firmware versions..The card used is Viofo 128GB MLC..



 

Attachments

  • 20190804_170745.jpg
    20190804_170745.jpg
    225.1 KB · Views: 8
  • 20190804_170902.jpg
    20190804_170902.jpg
    185.3 KB · Views: 7
Last edited:
Back
Top