VIOFO A119 V3

[Dash_vs_Evil]

Since I don't have my cams anymore, can the others with those problems tell if they use the official power supply/adapter or if it doesn't matter.
Another idea is that the ones without any issues tell us if they are using the official power adapter.
If so just try to run the cam from an integrated USB Port if available and check if there's any difference.
But still why is it that the issues (at least in my case) varied from FW to FW?
If it's the power adapter the problems should rather stay the same, right? No?

 

[vvs49]

below 12?

What is 12, I don't understand how it translates
In the photo, perhaps show these 12 !!

 

[Nigel]

What is 12, I don't understand how it translates
In the photo, perhaps show these 12 !!

The RGB pixel levels should be in the range 0 - 255, but they are actually in the range 12 - 255.

 

[vvs49]

Now I understand how and where you determineds?

 

[Nigel]

Now I understand how and where you determineds?

Not sure I understand the question, but I looked at the video you posted "A colleague from Romania sent me a video on my firmware, see if there is any flicker there ???", and saved a frame grab using MPC player, then looked at the values and histogram in a photo editor. Also looked at some of Dash_vs_Evil's videos and they are similar.

Possibly there are a few values less than 12, but if the exposure was changed so that 12 becomes 0 then I think the image would be better and the lines+flickering would be less visible. There are some players that incorrectly ignore the "Full Range" property and use 16 as the zero value, in which case the flickering would not be visible, I was thinking maybe the exposure was configured using one of those players!

 

[vvs49]

photo editor

That's not a camera issue, because In software like Davinci Resolve or video players these files looks identically. But in Premiere Pro, After Effects and Photoshop .MP4 files has wrong, shifted colors (looks like they're in Rec.601 space).
Did you see the shimmer there?
I have no time yet, I'll see later.
Still this can be from the 264 codec, it is for 1080P, and for 4K you need 265, I installed it myself by default.

 

[kwolf]

There's a lot of discussion about power supply in regards to the horizontal lines issue. For me, the problem didn't exist on 1.03 but it exists on 2.00 (perhaps related to the addition of HDR). Nothing changed with the power supply, so it would logically seem the issue is not with the power supply but with the firmware. I am using the official hardwire kit from VIOFO and my car has a good battery and alternator.

 

[jokiin]

Don't doubt there are improvements to be made in the firmware, but if you under power the camera it can and will cause problems, powering from built in USB ports is a poor choice, using the supplied power supply is preferred always

 

[kwolf]

I don't disagree - for sure it's better to use the supplied adapter. I just think if the same adapter is used with 1.03 and 2.00 and the lines only exist on 2.00, then the adapter in that event is not the issue.

By the way, the modded firmware someone posted here almost completely fixed the lines for me. However, parking mode is broken so I can't use it. If set to "Auto Event Detection" the camera will lock up after the first clip. If set to "low bit rate" it will continuously record at the same normal bitrate and quickly kill the MicroSD card. VIOFO just sent me the beta 2.1 so hopefully that works to fix the lines but doesn't have the same parking mode problem as the modded firmware.

 

[Kremmen]

I really am guessing, clutching at straws, but thinking about all possibilities.......

Some Ford windscreens, for example, have very fine, almost invisible heating elements. I wonder if the camera is affected by filming through this fine metal mesh.

I'd still be interested whether recording indoors, in the dark, from a mobile phone USB 'brick', away from the vehicle makes any difference.

 

[jokiin]

Some Ford windscreens, for example, have very fine, almost invisible heating elements. I wonder if the camera is affected by filming through this fine metal mesh.

that shows up as a slightly blurred line across the picture when you have that, wouldn't cause the flicker

 

[Kremmen]

Thanks, just checking.

How about the ferrite core that is on my Nextbase kit ? Would that help smooth the power, assuming we are still trying to rule power out or in.

One thing I have noticed is that the Nextabase HWK 1 is listed as incompatible with the Nextbase series 2 cams as it doesn't supply enough power even though both kits are 12/24v in and 5vdc out.

Nextbase post here:

Hi all, Just wanted to confirm this for you. The Series 1 Hard Wire Kit has two major differences to the Series 2 Hard Wire Kit. The Series 2 kit has a different size connector (which I can see some of you have found creative solutions for), but significantly, a different power output. The Series 1 HWK does not have adequate power output for a series 2 Dash Cam to fully function. The Series 2 HWK is however backwards compatible, so can be used on a Series 1 Dash Cam. Hope that helps. Regards, Millie- Nextbase

Technical Support Engineer
Nextbase UK https://www.nextbase.co.uk/

 

[jokiin]

I think sometimes people think all adapters/power supplies are created equally if the power ratings are the same, eg a nominal rating of 5v @ 2A or whatever (that's a reasonable assumption at face value) but aside from the quality of the power supply what you're powering makes a difference, we often get people wanting to power cameras off built in USB ports and telling us that the ratings are the same, it charges their phone etc no problem so why can't they, challenge is that things like phones may have 5v power supplies but if the voltage is a bit lower or sags under load etc it will still work fine for charging the 3.7v battery they use, the cameras with supercaps run 2 x 2.7v caps at 5.4v total, it doesn't take much of a voltage drop to cause stability issues, good stable power is important

How about the ferrite core that is on my Nextbase kit ? Would that help smooth the power, assuming we are still trying to rule power out or in.

no

 

[Nigel]

That's not a camera issue, because In software like Davinci Resolve or video players these files looks identically. But in Premiere Pro, After Effects and Photoshop .MP4 files has wrong, shifted colors (looks like they're in Rec.601 space).
Did you see the shimmer there?
I have no time yet, I'll see later.

In a Rec.601 file, the visible values start at 16, it is OK for dark noise to be recorded in values 12, 13, 14, 15 because they will all be displayed as pure black unless edited. So someone seeing this file displayed as Rec.601 would not see most of the flicker and lines. However the file is marked as full range, the visible values should start at 0, so when displayed correctly as a full range file, noise at level 12 will be easily seen.

My concern is that whoever calibrated the exposure set it up using a Rec.601 display and not the full range display they should have been using, that would explain people seeing the noise when they play it properly. If it was calibrated correctly using a full range display, why are there no values less than 12?

Yes, some software will display or edit it incorrectly, and some operating systems/graphic cards will be configured incorrectly, and some monitors are incorrectly calibrated. There are many reasons why different people will see different amounts of flicker/lines noise. My question is only, why is none of the image darker than level 12 when it is a full range file that should go down to 0?

Still this can be from the 264 codec, it is for 1080P, and for 4K you need 265, I installed it myself by default.

I have never tried the A119 V3 configured for 4K and H265, what colour space does that use? Maybe that has no flicker?

But the raw files we are looking at and which have some visible flicker are all H264 and marked as full range BT.709. If they started at 0 instead of 12 then I don't think anyone would be complaining about flicker since it would be nearly invisible even on the noisiest camera!

 

[Kremmen]

@Nigel - Can you give my file a check to see what RGB I'm collecting:

>> Krem Sample <<

 

[Nigel]

@Nigel - Can you give my file a check to see what RGB I'm collecting:

>> Krem Sample <<

That goes down to zero, nice exposure, but it was daylight so a different situation, might be a different result in full darkness with headlights only.



(Shape of the graph is not significant since it depends on the scene contents, only the spread of values which does go the full range from 0 to 255, with all the sky overexposed at 255 as is common on the A119 V3!)

 

[Kremmen]

Night sample ...... your wish is my command :

>> Krem Night Sample <<

Thanks for your time

 

[Nigel]

Night sample ...... your wish is my command :

>> Krem Night Sample <<

Thanks for your time

That also goes down to zero, nothing wrong with it.

So why is your's good and noise free when the Dash_vs_Evil videos are wrong and have noise?



Might be nice if yours could be brightened a bit in the midrange but that is probably not easy to do with auto exposure:



Instead of:

 

[Kremmen]

Isn't this very odd.

Have Viofo or Sony changed something ?

 

[petrlustich]

It looks to me like your processing before giving it to youtube may be partially responsible...

I did not compress it before uploading.