A119 vs A119S Video Comparison - Quick Analysis

AndrewL

Active Member
Joined
Aug 10, 2015
Messages
160
Reaction score
208
Country
Canada
I’ve directly compared the A119 (firmware version v2.0) and the A119S (v1.0 production). It’s quick and there’s more work to be done.

The A119 has better video quality in the majority of situations both day night. It’s sharper and has less motion blur.

The A119S has much less distortion in the corners due to the reduced field of view. It has slightly better dynamic range at night, but it’s not a significant improvement even under very low light. The A119S has to be better in most aspects than the A119 to even warrant consideration as its $20 more and I’m not seeing it.

The A119 at 60FPS runs close to natively and there’s not much reduction in quality compared to other cameras. However the A119S has 12% more pixels to record and a 16% reduction in bitrate compared to the A119 (20mbps vs 24mbps respectively). The A119S at 30FPS is 15mbps which is more appropriate and should be an improvement in situations with a lot of details. More testing to come.

From what I understand the A119S’ Sony Exmor IMX291 firmware is still new and image improvements are coming. As well I want to do a number of tests I listed at the bottom. Full review out later. I can post raws if there are requests for specific footage.

Here’s a video combining a few clips, make sure you watch it in 1440P.


As well here are some stills that you can find on IMGUR:


Shows the distortion of the A119 lens towards the corners


A119 is much better otherwise, sharper at much further distances

Overall sharper at night






60 FPS

I also want to bring attention to claims that 60FPS reduces motion blur or are sharper. That is not true. It’s shutter speed which determines how sharp an image is.

60 FPS means each second is divided into 60 frames. Each frame has a max shutter speed (how long the camera lets in light) of 1/60th. However the shutter speed can be much faster let’s say 1/5000th of a second every frame.

This is why during the day if you freeze frame 30FPS video, the license plates are sharp because the shutter speed is very fast. At night there is less light so the camera can compensate by either slowing the shutter speed to increase light or bump up the ISO which increases noise.

Here’s a good image which explains what I mean:
11124d1236218677-progressive-shutter-speed-frames-rate-fps-chart.jpg



Further Testing
I need to test both camera on both halves of my windshield and on a larger vehicle. This can help rule out distortion and other effects which can cause problems. I want to do some more dynamic range testing which should play into the A119S strengths. The daylight video was overcast, not many shadows.

I also want to more thoroughly test out 60FPS mode.


UPDATE: New Video

Did some comparisons on the left hand side which appears less blurry. A119S looks better and more on par with the A119. The corners the A119S definitely has a lead. In other parts the A119 is sharper. According to the owner of Street Guardian says the firmware has a ways to go so it's possible in the future the video quality will be improved.

Right now I do think the A119 remains better but that may change in the future.
 
Last edited:
Thank you very much Andrew for your review. It means a lot! I've been experiencing exactly what you are describing in terms of quality and sharpness on A119S

Edit: I'm sure your windshield is super clean, what is your experience with glares at dusk/nighttime?
 
Thank you very much Andrew for your review. It means a lot! I've been experiencing exactly what you are describing in terms of quality and sharpness on A119S

Edit: I'm sure your windshield is super clean, what is your experience with glares at dusk/nighttime?

You're welcome! Windshield is semi clean because of the salt spray from the road. Glare has been reduced quite a bit because I've thrown a black towel which absorbs a lot of the reflection so I actually don't know how bad it is. I have the CPL so I'll probably run some tests later on.
 
You're welcome! Windshield is semi clean because of the salt spray from the road. Glare has been reduced quite a bit because I've thrown a black towel which absorbs a lot of the reflection so I actually don't know how bad it is. I have the CPL so I'll probably run some tests later on.

I apologize I wasn't clear. I was wondering what's your feedback on the street lights and oncoming vehicle light glares.
 
Halo on street lights i think is what TekkieNYC is thinking about

index.php
 
I've observed on many occasions that if your windshield is super clean, the flaring/halo that the camera captures is almost or exactly the same as what the human eye can see through the same windshield.
 
However the A119S has 12% more pixels to record and a 16% reduction in bitrate compared to the A119 (20mbps vs 24mbps respectively).
Why does the A119S have 12% more pixels to record in your example of 60fps at 1080p? Maybe I misunderstood you.
 
Are both videos recorded at 30fps for the comparison?
 
Why does the A119S have 12% more pixels to record in your example of 60fps at 1080p? Maybe I misunderstood you.
Lower resolution but twice as often.

1440p is 3.6MP
1920 x 1080 x 2 = 4.1MP
4.1 / 3.6 = 1.12

Video encoding works with differences between frames so the data rate won't go up linearly as you take frames more often, it's probably log or something.
 
  • Like
Reactions: Mtz
I apologize I wasn't clear. I was wondering what's your feedback on the street lights and oncoming vehicle light glares.

Looks alright to me, I don't have the most scratch free glass. Also, it hasn't been something I've been analyzing so I don't have a good basis for judging. Can you tell me what you're looking for and how it affects video quality so I can do some deeper probing?


Lower resolution but twice as often.
1440p is 3.6MP 1920 x 1080 x 2 = 4.1MP 4.1 / 3.6 = 1.12

Video encoding works with differences between frames so the data rate won't go up linearly as you take frames more often, it's probably log or something.

Hoping someone who is more knowledgeable on encoding can chime in. Good point you made.
 
Hoping someone who is more knowledgeable on encoding can chime in. Good point you made.
Depends on the amount of movement in the image, if there is no movement then 60fps will probably only need the same bitrate as 30fps, the full frames will occur at the same rate in both, only the differences/movements will occur twice as often. If there is a lot of fast movement then 60fps will require a lot more than 30fps but nowhere near double.

On you frame grabs, the A119 image appears to be at native resolution, but the A119S image has been enlarged and is interpolated, the cars in the A119S images are actually larger than in the A119. How did you enlarge it? Might of been better to have shown both at native resolution with black bars down the side of the A119S image so that the A119S isn't blurred by the interpolation...

Interestingly, even though the A119S has less pixels in the sensor, it appears to show more texture detail in the road, and the cracks in the tarmac have more detail, it also gives more detail in the clouds and lights that are overexposed in the A119, probably means that license plates in sunlight and when reflecting headlights will be better on the A119S?

I can't see any real motion blur, the wheels on the cars which should be the fastest moving things are all sharp on both cameras even at night, we could do with some high speed images to compare the motion blur since that should be the main advantage of the A119S.

The A119S tends to be more blurred than the A119 on the right, but it is the other way around on the left. Have you checked them without a windshield in the way?
 
On you frame grabs, the A119 image appears to be at native resolution, but the A119S image has been enlarged and is interpolated, the cars in the A119S images are actually larger than in the A119. How did you enlarge it? Might of been better to have shown both at native resolution with black bars down the side of the A119S image so that the A119S isn't blurred by the interpolation...

Interestingly, even though the A119S has less pixels in the sensor, it appears to show more texture detail in the road, and the cracks in the tarmac have more detail, it also gives more detail in the clouds and lights that are overexposed in the A119, probably means that license plates in sunlight and when reflecting headlights will be better on the A119S?

I can't see any real motion blur, the wheels on the cars which should be the fastest moving things are all sharp on both cameras even at night, we could do with some high speed images to compare the motion blur since that should be the main advantage of the A119S.

The A119S tends to be more blurred than the A119 on the right, but it is the other way around on the left. Have you checked them without a windshield in the way?

Didn't see the interpolation when I watched the YouTube video. The A119S cars are larger because it has a smaller field of view and there's some distortion going on in the corners.

Regarding the texture in another review between the Blackvue DR650 vs Thinkware F77, the blackvue was better in that way too. However looking at the license plates the F770 kicked its butt. I don't know the reason but yes, the ground does look a lot better.

It's the license plates I think that are of greatest concern with details. The face second, I need to record more videos.

Yes I agree the a119s does more blur on the right side and I'm planning on shifting the cameras to the other side, possibly creating a rig in order to walk down the street with the two cameras with nothing in the way.

There's a lot more testing to be done but at the end of it I don't see how the A119S will beat the A119 which it needs to do to justify its higher price tag. I'm hoping for the image improvements, I'm not an expert in firmware so I don't know how much that will help.


Edit: Enlarged it through Adobe premiere. Had it expand the image, cropped it and I screen grabbed a jpg.
 
There's a lot more testing to be done but at the end of it I don't see how the A119S will beat the A119 which it needs to do to justify its higher price tag. I'm hoping for the image improvements, I'm not an expert in firmware so I don't know how much that will help.

the OV sensor has had a one year head start on this platform so will have an advantage right now, that will change as further work is done with the Sony sensor I'm sure
 
@AndrewL Can you please tell me this.... with future firmware updates in your opinion what can an update really do OR not do. Being not so informed on the benefits of an firmware update can the A119S be totally improved (saved) from the condition it is now in, or is there only so much that can be done to make it sharper no more blur etc. In other words can the A119S show as well as the A119 which in my opinion is a great DVR . I know the A119S has great hardware so I guess its just up to the firmware now. Am I correct to think this?
 
the OV sensor has had a one year head start on this platform so will have an advantage right now, that will change as further work is done with the Sony sensor I'm sure
I hope you're right. I started off with a single A119 and then waited until the A119S was released to pickup 2 more for the other cars that are mainly driven at night.
 
@AndrewL Can you please tell me this.... with future firmware updates in your opinion what can an update really do OR not do. Being not so informed on the benefits of an firmware update can the A119S be totally improved (saved) from the condition it is now in, or is there only so much that can be done to make it sharper no more blur etc. In other words can the A119S show as well as the A119 which in my opinion is a great DVR . I know the A119S has great hardware so I guess its just up to the firmware now. Am I correct to think this?

Oh man, my reviews would be so much better if I had the engineering and programming knowledge to answer this question. I think @jokiin and @viofo would be much better informed to deal with this question. I am also very interested as well.
 
Now that both A119 and A119S are in stock on the new OCD Tronic Amazon store this week, I'm starting to get questions asking what the difference between A119 and A119S is.

The basic quick summery I give is

A119S =
1) Native Sony 1080P sensor so larger pixels should result in better night performance (once firmware is tuned up further)
2) 60 FPS at full res
3) 7G F1.6, 135° FOV (less wide, bigger aperture)
4) CPL Support

A119 =
1) Higher Resolution OV 1440P sensor but smaller pixels, so less light available at night
2) 30 FPS at full res
3) 6G F1.8, 160° FOV (wider, smaller aperture)
4) Newest production also has CPL Support

I look forward to digging into the finer technical comparisons as people do more side by sides like this. (and A119S firmware IQ matures further)
 
When we count number of glasses in the lens, does that included IR filter?
 
3527228.jpg

How mush will the different in aperture change in the depth of field.
Or maybe I should ask in this way. Is that the reason the A119 in focus at longer distances?
 
Last edited:
3527228.jpg

How mush will the different in aperture change in the depth of field.
Or maybe I should ask in this way. Is that the reason the A119 in focus at longer distances?
The narrower field of view probably makes more difference, but it is not going to be noticeable if the lens is correctly focused. Just means that the A119S lens needs to be focused slightly more accurately.
 
Back
Top