A129 Pro firmware

[EGS]

I'm rather pissed that the A129 Pro did not record an incident today. The 5 minute file that should have contained the incident was only 45 seconds long, and ended before the incident. The next file started 5 minutes after the previous file, so 4 minutes and 15 seconds was missing.
I did not hit the save on my Tesla because I assumed that the A129 Pro caught the incident, and by the time I had checked the Tesla data, more than an hour had gone by, and it had been overwritten.

If you had an impact, then the camera should have saved that event in to the RO folder.

 

[DAP]

If you had an impact, then the camera should have saved that event in to the RO folder.

I was not involved in the incident. A motorcycle turning left in front of me from a cross street (not doing anything unsafe) fell over.

 

[EGS]

I was not involved in the incident. A motorcycle turning left in front of me from a cross street (not doing anything unsafe) fell over.

I see now. It sucks the camera decided to fail at the least opportune moment.
What firmware version are you using? Maybe we testers need to review our footage with more frequency to see if we find missing footage and see if that is sd card related or if there is a bug affecting a particular revision of the firmware.
I hope there is a way to figure out what caused the error so it can get addressed.

 

[R1KSU]

Go to your movie files, at the bottom is a folder called RO, your files should be in there, just search for date and time your missing file, should be

 

[Gordohat]

Maybe we testers need to review our footage with more frequency to see if we find missing footage

They should put the double beeps back in the code for when there is a loss of recording. I think Viofo want to hide the frame loss issues rather than address it.

 

[Vlad1977]

There are two things here...first is if the camera should have locked the footage...the second is if there is missing footage.

If there was no impact, then the file will not be in the RO folder. How does the camera know whether to lock any footage? Only if the g-sensor / impact occurred. If that's not the criteria, then all of your footage would be locked and go into the RO folder. Manually saving the footage would cause the footage to be placed in the RO folder.

Missing footage is either because it was not recorded or it was recorded over. From DAP, it sounded like he wasn't involved...no impact...and there was a missing file for the footage expected. However, DAP indicates an hour had gone by and the file was overwritten. It can't be both. If it was overwritten, then it's not missing. You can assume that it wasn't overwritten if you remember the time it occurred and you can find it by going through the files on the card. If the file with the time you're looking for is not there and there are no gaps in the sequential naming of the files, then it was overwritten. If there is a gap, you need to review the RO folder to see whether or not it's there.

I have thought about the issue of files being overwritten. I've decided to buy as large a card as I can...when I can ...maybe 512GB, to handle this as much as possible. There are factors like bitrate, video quality/resolution, etc, that factor in to how big a card is needed. I went on a trip and wanted to look at some points of interest but my camera had overwritten the footage on my 64GB card. Other factors to consider is the number of minutes set for each video slice, the amount of space reserved by normal video and the amount of space reserved by locked footage. I believe I read 80% is reserved for normal footage and 20% for locked footage. The idea of the camera is to capture important events like impacts. If you don't access the footage immediately, then you can assume that it wasn't that important. The video of your drive to work and home everyday is not that memorable and so if it's overwritten, no big deal. The camera doesn't know if something is important unless it's an impact or you manually tell it to save the footage. I've dropped down to a lower resolution, for now, so the card can store more video before being overwritten...those big cards are expensive.

For parking, the settings for capturing video are impacts as well as motion as those are relevant for a parked car. Parking settings also allow setting it to always record at a low bit rate with very small files. I have mine set to low bitrate video.

As for "missing" footage, I think I'll have to pay closer attention to see if that's happening. I've been assuming it hasn't, but you can only be certain if you check it. Mine does the beeping when the g-sensor believes something happened. I set it to be less sensitive as I was getting beeps going over potholes and bumps and such. I was hoping the firmware could connect to my router so I could transfer and view video from my PC without having to remember to take out, transfer and then return the card to the camera. I asked on the VIOFO forum but haven't got an answer from VIOFO about this feature in the latest firmware for the A129 Pro Duo.

So, I'm thinking that in DAP's case, perhaps the video file was overwritten and nothing really failed.

 

[Gordohat]

perhaps the video file was overwritten and nothing really failed

That is not how a file system works. If the file was overwritten, you would not be seeing any part of that old file still addressable from the file system. You would need to forensically examine at the bit level.

 

[jokiin]

put the card back in the camera and go to playback mode, if the file is there but broken the camera will repair the file, if that doesn't happen then the segment isn't there, maybe not recording at the time

 

[programatix]

From DAP's post, it seems that he has 2 devices that are recording. One is the A129 Pro and the other is his Tesla (the car?). He didn't initiate manual save using Tesla because he thought that the A129 has also recorded it. After he found that the A129 recording is corrupted, he also found that the Tesla has overwritten its recording as it has past the 1 hour limit.

His complain is on A129 Pro saving corrupted video.

2 things here. Firstly, how can a dashcam that supposed to record failed to record? Secondly, he was not aware of this as A129 Pro latest firmware remove the beeping notification when write error occurred, else he would have perform the manual save on his Tesla.

BTW, pressing the emergency button on the A129 only move the current recording into the RO folder. So if the file recorded is already corrupted, pressing the button would just move that file into the RO folder.

Sent from my SM-N975F using Tapatalk

 

[EGS]

They should put the double beeps back in the code for when there is a loss of recording. I think Viofo want to hide the frame loss issues rather than address it.

I totally agree with that.
The fix they implemented is similar to someone applying a piece of electrical tape on top of the check engine light on a vehicle to ignore that annoying warning.

 

[EGS]

Another possible scenario is if there was a momentary glitch in the clock and consequently it would break the file sequencing causing for the file to be placed out of order. If that file went to the top or the beginning of the sequence, then that file likely got overwritten not long after the camera continued recording. Something to consider looking at.

 

[SawMaster]

Another possible scenario is if there was a momentary glitch in the clock and consequently it would break the file sequencing causing for the file to be placed out of order. If that file went to the top or the beginning of the sequence, then that file likely got overwritten not long after the camera continued recording. Something to consider looking at.

Anytime you have a lost or corrupted file, study the file structure for date/time, size and numbering. Often there will be a clue in there pointing toward the problem

I always keep a spare card on hand. Rather than trust a cam to lock a file I leave the cam going for longer than my chosen file length time, then I do a manual shutdown to swap cards. Seems to be the most positive way to ensure you'll have the recording you want with any cam. You're going pull the card when you get home anyway so no real loss of your time in total. Nothing is more aggravating than not having a file you thought you had

Phil

 

[kamkar]

For many years i just used a mental note if there was something i wanted to save, this have cost me many things over the years, but always due to bad memory on my part not overwriting.
I now use the manual event button, and only retrieve those files in the RO folder every 2 weeks or so.
I do also carry a spare card in the car, but it is mainly used for when i have had a card out but then forgotten it at the computer, and even if i have several cameras on my windscreen i dont like if one of them are not recording.

 

[DAP]

I'm using a 256GB card in the camera, this should record more than 10 hours of video.
I have the camera set to 5 minute files so that they don't exceed the 4GB limit of FAT32.
There are no missing files.
The file that should have contained the event was under size and contained only 45 seconds.
The next file started 5 minutes after the short file started.
The 1 hour limit applies only to TeslaCAM (something I REALLY wish that they would fix).

edit: I decided to take a closer look at the files. It looks like every file was being truncated to 45 seconds that day. The time between files was random, sometimes only 5 seconds other, as in this case, several minutes.
In the file listing below, the middle file is the one that should have contained the incident.

-r--r--r-- 1 DAP DAP   49114547 Nov 27 06:06 20201127_140558_00504PF.MP4
-r--r--r-- 1 DAP DAP   76266079 Nov 27 06:07 20201127_140704_00505PF.MP4
-r--r--r-- 1 DAP DAP   64221779 Nov 27 06:11 20201127_141045_00506PF.MP4

Earlier that day, there were some 1.9GB 5 minute files. but during the drive where the incident was, they were all about 64 MB.

I have an identical rear camera with exactly the same type of memory card. The rear camera was recording perfectly at the time, but because it was pointed to the rear, missed the accident entirely.
Both cameras are running version 2.1 with build 20200803

edit2: I just drove for 300 miles. The front camera worked properly for 125 miles until I stopped to supercharge. After that, it only recorded 45 second files. I had formatted the card in the camera before starting, so the memory card is not full. The back camera worked perfectly. It recorded 59 second files while parked at the supercharger. and always 5 minute files while moving.
I'm beginning to suspect that the problem has something to do with parking mode. It is like the front camera gets stuck in parking mode, and does not notice when the car starts moving.
I am going to move the front memory card to the back and the back to the front. I will go through the settings and verify that both cameras are set up identically. I will format both memory cards in the camera that they are installed in.

edit 3: Parking mode was off on the rear camera, but it behaved as if parking mode was on.

 

[DAP]

Since my last post, the Parking mode was off on both cameras, and both cameras are working properly, even the motion detecting parking mode (Apparently, that parking mode switch only applies to the hard wired parking mode. I only have power to the cameras).

I can't say that I have driven much since then, I certainly have not filled the memory cards yet.
Other things I noticed: The motion detecting parking mode does not seem to have any pre-run buffer. While parked in a parking lot, the parking mode videos started when passing cars were already in front of the camera, and the camera caught them going away.
I was impossible to read license plates of oncoming cars on a two lane road in broad daylight (near noon). This is with 4K video on a 49" 4K monitor.
Video from the front camera was over exposed, the back camera was correctly exposed. Both cameras were purchased at the same time. I have since lowered the exposure on the front by one stop.
The back windshield was interacting with the polarized light from the sky and the CFL filter in such a way that I could tell the location of the sun in the sky, even if it was not in the picture. The patterns of light and dark would rotate as I turned the corner.

 

[Nath]

... While parked in a parking lot, the parking mode videos started when passing cars were already in front of the camera, and the camera caught them going away. ...

While still in parking mode? I saw similar behaviour except it was caused by the fact (or at least i thought so) my unit switches from parking mode to normal mode as soon as motion is detected (for God knows reasons). So as it should, about 5 seconds is lost during switching process and no buffer.

I dumped my two A129Pros. Tried DR900X, disappointedly returned it. And now i'm waiting for A139 and i hope for something interesting.

 

[DAP]

The parking mode was turned off, but motion detection was enabled for parking mode. I need to go through and document all my settings. Some are not very clear what they do. With the setup as I had it, it would record continuously when I was driving, but only record when motion was detected when the car was parked. The camera is connected only by a 2 wire power connection to a USB charger powered by the cigarette lighter port in my car. The camera itself is decideding if the car is parked or not.

 

[programatix]

I recently noticed that my A129 Pro Duo is behaving weirdly every time when switching from parking mode to driving mode. Instead of normal switch, it reset itself. Sometimes it reset 2 times before being able to enter driving mode.

I already tried formating the sdcard and the problem remain. This actually started after a long journey I made (8 hours nonstop). After that, the next day when I start the car the problem was seen and remained since.

Anyone encountered such problem before?

Sent from my SM-N975F using Tapatalk

 

[Mtz]

Change the power source, do not use the mount power connector, try with another card.

 

[LeJerk]

People should really start separate threads for all these different issues and use this one just for talking about the firmware. It's gone way off track it seems like.