VIOFO A139 Pro - Testing / Review Info

[rcg530]

I've asked for samples of the newly announced BlackVue dash cameras. In theory, I'll be getting some to review. The spring availability of the DR970X will certainly delay that comparison.

 

[TonyM]

Resetting the firmware settings to their defaults with firmware version 1.0_1217 and leaving the HDR setting at "On" rather than "Auto", everything seems to be working now. I've sent an email to VIOFO about the behavior I encountered.

My advice would be to not use the A139 Pro HDR "Auto" feature just yet.

I had the same problem after trying HDR Auto with FW V1.0 1217. The camera just beeped constantly after booting with the REC light flashing. I too have set my A139 Pro to HDR-ON.

 

[Vortex Radar]

k, I'm working through some A139 Pro footage now while working on a Starvis 2 video. Here's a few preview images:

Driving into the sun with HDR on:



A129 Pro vs. A139 Pro, rear plates with headlights, both dashcams with HDR/WDR on:



another similar frame grab:



A139 Pro, HDR on vs. off



Those are stationary samples. I'm wanna look at some moving ones too.

 

[Panzer Platform]

k, I'm working through some A139 Pro footage now while working on a Starvis 2 video. Here's a few preview images:

Very nice. That screenshot driving into the sun looks good.
Is the A129 Pro IMX317 really a STARVIS sensor?
I remember reading a post from @Mtz and he said it wasn't a STARVIS sensor but a CMOS Exmor R Sensor.
Looking at the data sheets, the IMX317 does not have the STARVIS logo.
Maybe I'm looking at the wrong datasheet.

https://www.sony-semicon.com/files/62/pdf/p-13_IMX317CQC_Flyer.pdf

https://www.sony-semicon.com/files/62/flyer_security/IMX678-AAQR1_Flyer.pdf

Oh yeah, do you have both cameras in 1-channel configuration?
@rcg530 discovered the night time HDR capability can drastically be reduced when running the A139 Pro in 2&3 channel configuration due to the camera automatically reducing the Bitrate.

 

[Dashmellow]

I remember reading a post from @Mtz and he said it wasn't a STARVIS sensor but a CMOS Exmor R Sensor.

STARVIS is apparently part of the Exmor R and RS family of sensors, so they are closely related.

"Sony STARVIS sensors have the same architecture as that of Exmor R sensors, only that they offer better image quality in extreme low light conditions with enhanced NIR sensitivity."

(from Wikipedia - Exmor)

"STARVIS is a series of sensors with the Exmor RS family. It features high pixel sensitivity, making it suitable for low light applications. Industrial versions are finding applications in ambient-light surveillance systems. Commercial versions are finding applications in prosumer webcams with 4K HDR support, based on single exposure."

See also:

Sony Exmor vs Sony STARVIS sensors; a detailed comparison - e-con Systems

Exmor, Exmor R, STARVIS, and Exmor RS are four highly popular series of Sony sensors in markets like industrial, retail, medical, etc. Get a detailed comparison of these sensors in terms of their structure and architecture.

www.e-consystems.com

"Sony STARVIS is a series of CMOS sensors within the Exmor R family that offers a sensitivity of 2000 mV/μm2 or more. It is also a back-illuminated pixel technology developed for CMOS image sensors designed to deliver a high image quality in visible light as well as NIR (Near-InfraRed) light regions.

Though the Exmor R series of sensors came with better sensitivity and lower noise, their performance in the NIR spectrum was not as good as the previous generation of sensors. The Exmor RS family overcame this challenge by increasing the pixel well depth. In addition to this change, Sony introduced a new architecture in Exmor RS called the stacked image sensor architecture. In this new design, the sensor circuitry for each pixel is arranged below the silicon substrate instead of beside it. This also helped in collecting more light in the NIR region, thereby increasing the QE (Quantum Efficiency) in that spectrum."

 

[Panzer Platform]

Starvis is apparently part of the Exmor R and RS family of sensors, so they are related.

Thanks for the clarification.
I read the links, now my brain hurts, and the new info pushed out what I remembered about multiplying fractions as a child. lol

 

[Vortex Radar]

Ah yeah, looks like you're right. I didn't realize that Starvis was a subset of Exmor and that the names weren't interchangeable. My mistake! Thank goodness we're talking about it here before I made a mistake in the YouTube video, lol. I can switch to comparing the A139 vs A139 Pro since those really are Starvis 1 vs 2. Resolution is obviously different, but yeah.

Since I'm working in my video editor now, here's another quick sample. Looks like HDR in the Starvis sensor is more effective than WDR in the Exmor sensor. Better dynamic range and less noise, plus no pink cast like WDR adds.



As for the bitrate thing, yeah that's annoying so I've been running the A139 Pro (and other dashcams too) in 1CH mode for all the video quality comparisons like this.

 

[Panzer Platform]

I can switch to comparing the A139 vs A139 Pro since those really are Starvis 1 vs 2. Resolution is obviously different, but yeah.

I was struggling to decide which camera to compare with the A139 Pro for my HDR test footage in my YouTube review, (A129 Pro, or A229).
So I just did all three. lol

 

[Vortex Radar]

I was struggling to decide which camera to compare with the A139 Pro for my HDR test footage in my YouTube review, (A129 Pro, or A229).
So I just did all three. lol

lol...

Yeah I'm having a heck of a time with testing configurations. Multiple cameras, HDR on vs. off, WDR on vs off, CPL on vs. off, etc.

 

[Panzer Platform]

Yeah I'm having a heck of a time with testing configurations. Multiple cameras, HDR on vs. off, WDR on vs off, CPL on vs. off, etc.

I know right?
I was so conflicted on the issue of settings.
So I decided to try to use as many default settings as possible since the average consumer with most likely plug it in, set it, and forget it.
So I went with HDR "ON" for the A139 Pro (default).
And WDR "OFF" for the A129 Pro & A229 (default).
And in the interest of continuity from all my previous reviews I only use the CPL during the day, and remove it at night.

 

[LateralNW]

@Panzer Platform @Vortex Radar
And firmware updates that can change the way previous functions work.
Usually not mentioned in the change log.

 

[Vortex Radar]

@Panzer Platform @Vortex Radar
And firmware updates that can change the way previous functions work.
Usually not mentioned in the change log.

Oh it makes it even more challenging to make videos when the more time goes on, the less valid the comparisons become. Such is life with tech products.

 

[vvs49]

Gentlemen!
Do not forget that the A139 firmware with HDR is still imperfect compared to A139 Pro, I'm waiting for corrections from VIOFO or I'll fix it myself, there is no weather!!

 

[Vortex Radar]

Gentlemen!
Do not forget that the A139 firmware with HDR is still imperfect compared to A139 Pro, I'm waiting for corrections from VIOFO or I'll fix it myself, there is no weather!!

Correct it’s currently still using the same HDR implementation as Starvis 1. I’m wanting to demo the improvements of Starvis 2 over 1, but it’s harder without that. I do see less noise than the A129 Pro with an Exmor sensor, but more similar to the A139 with a Starvis 1 sensor.

 

[rcg530]

• Problem
  • A139 Pro v1.0_1217 firmware
  • HDR = Auto
  • HDR Time = On 18:00, Off 08:00
  • A139 Pro freezes or fails to boot successfully (beep-beep)

I reported this issue a couple of weeks ago. The Android VIOFO app v3.2.9 presented new options for the HDR feature (HDR timer feature just like the A119 Mini). When you set the "HDR" setting to "Auto" and set the new setting of "HDR Time" to "On 18:00, Off 08:00", it results in an unstable A139 Pro and/or it would fail to boot successfully (starts to beep-beep continuously). VIOFO reached out to me after the CNY break looking for footage to use in a marketing video for the A139 Pro. I inquired about the status of this issue. I was told they could not reproduce the problem.

I setup a test in my filming studio using an A139 Pro 3-channel with firmware v1.0_1217. I set the HDR settings as mentioned above. I was able to reproduce the problem and I captured video footage of the problem. I put a video together showing the VIOFO App screen along with video of the A139 Pro. I submitted that to VIOFO.

This morning, VIOFO stated they were able to reproduce the issue I show in my video. VIOFO is working on a fix and hopefully it will be in the next firmware release for the A139 Pro.

If you want to see the video I submitted to VIOFO, it's on my Google drive. At the 5:25 in the video is when the problem first shows up during a power up sequence. The mp4 file is 503.4 MB.

a139pro_10_1217_hdr_auto_problem_public.mp4

drive.google.com

 

[Vortex Radar]

More screenshots.

Opposite direction plates in the shade, Starvis sensors do a better job of freezing motion.



More examples, but all plates blurred.





Surprisingly, I'm not seeing the double exposure effect with the A139 Pro with HDR enabled the same way I'm seeing with the A139. I don't believe that Viofo has enabled Clear HDR on the A139 Pro yet.



Parking garage test, A139 Pro does a better job of freezing motion again.



and like with the headlights shining on plates in front of me, HDR again does a much better job of capturing highlight detail than WDR.

 

[Panzer Platform]

Wow, the "standard" A139 is hanging in there, I wonder if it's going to fall apart at night.
The last screenshot with the A129 Pro garage doorway is yucky. lol

 

[kamkar]

That is what i have been thinking, ATM the cameras do traditional HDR bundling consecutive frames, CUZ Sonys own Clear HDR so not have the same look.
I am wondering of the associated SOC in the cameras can even do the clear HDR or if it is not up to the SOC tell the sensor to do clear HDR

I think so as the HDR footage i have seen so far have this look of a smeared ( motion blur ) slow exposure + a darker faster exposure frame mixed together.
But Sonys own material on clear HDR do not have this look, its just razor sharp ( capture of a pendulum in motion )

 

[serginho]

@VIOFO-Support
Hello Viofo,
I have model A139 Pro and I am happy about it, but I have one request if possible.
Whenever Parking mode is in "Auto event detection", there are 2 settings : Parking G-sensor and Parking Motion Detection (options: Low, Medium, High).
Is it possible to have also option "Off" for Parking Motion Detection so that only G-sensor to trigger ?
I always have not very useful videos from motion detection and I am only interested if someone hit my car.
Thank you in advance!

 

[mj1985]

@Vortex Radar thanks for all the screenshots. Looks like the A139 Pro can solve my problem with over exposed license plates with my Matrix LED light.