A129+ new user... finally

[ursamajor]

it's correct, just irrelevant

I give up...

...but it's good for me to know that SG advertise correct FOV. I'll take a closer look on. I even wanted to buy a SG after the masquerade with that returned V3, but I noticed that in RO are sold only 1080p models and I wanted at least 1440p (QuadHD).

 

[ursamajor]

Though, I also have to say that I'm enough satisfied with my A129 Plus. The pluses are more than the minuses at this camera. I can live with 120 deg. H-FOV instead of declared 140. In fact, I think a value of 120-130 deg. combined with the QHD sensor (even better with a 4k) is ultra-sufficient for a dashcam. IMHO, it offers the best compromise between wide field (to capture more action on the road) and good readability of the plates.

 

[jokiin]

Though, I also have to say that I'm enough satisfied with my A129 Plus. The pluses are more than the minuses at this camera. I can live with 120 deg. H-FOV instead of declared 140.

I notice they just call it a 140° Wide Viewing Angle, they don't specify that as a horizontal value, regardless of the numbers if you are satisfied with it is the main thing anyway

 

[ursamajor]

I notice they just call it a 140° Wide Viewing Angle

Here we go again! At V3 they also claim 140° Wide Viewing Angle, and be it either horizontally or diagonally, it's wrong/lying. At Plus it's only closer to advertised.

 

[jokiin]

Here we go again! At V3 they also claim 140° Wide Viewing Angle, and be it either horizontally or diagonally, it's wrong/lying. At Plus it's only closer to advertised.

yeah here we go, missing the point again obviously, you say you can live with your 120° horizontal viewing angle even it's not the 140° that they advertise it as, problem is they don't advertise it as a horizontal viewing angle anyway, that was your assumption

 

[ursamajor]

yeah here we go... problem is they don't advertise it as a horizontal viewing angle anyway, that was your assumption

Ignore the 120° horizontal of my Plus (it's too close to 140° and confusing) and concentrate on real 95° horizontal FOV of V3 vs. 140° advertised. It's not true in any of the cases: horizontally or diagonally.
You cannot convince me to "swalow" a lie, even packed in good tasting glaze. That's the point you are missing.

I'm out ot this. As a manufacturer, of course, you'll never accept the reality (though you know it well), which is one and is related to true/real FOV of a product/camera. You can fool the ignorant customers, but that doesn't mean you're right.

PS: one more thing... the real FOV is so easy to be measured. They have a bunch beta testers, they could ask them to measure the real angles and use the measurement in advertising the product. But they still prefer to advertise lies. What we are talking about?

 

[jokiin]

is the plus really 120° horizontal, that seems way wider than what they normally use

and concentrate on real 95° horizontal FOV of V3 vs. 140° (unspecified) advertised. It's not true in any of the cases: horizontally or diagonally.

the 140° may well be a real spec, just not one that you are able to measure, to give you some context on what I mean in a previous model camera we did we used a lens which gave a 112° horizontal field of view, the diagonal field of view was 131° when used with the sensor we were using, from the manufacturers spec sheet the lens was a 152° lens, I expect a lot of manufacturers would advertise that as a 152° wide angle lens, and technically it was, we never advertised it as such

I'm not accepting your reality as you are assuming what the numbers quoted are specifying when they don't actual make any mention of what the figure is meant to indicate, and you don't actually know, you have just assumed what it means, you're putting words in their mouths, I'm not suggesting that that's how things should be where people are left to assume or guess or whatever, it's confusing and I get that

 

[SawMaster]

Street Guardian is the only dashcam manufacturer whose FOV specs reflect the truth. They might be given diagonally but they are definitely not over-rated just to gain sales. Jokiin and SG are honest about their cams.

Specs and features of dashcams have always been stretched to the breaking point in advertising. It's not going to change with everybody else doing it. I've seen cams advertise a 170 degree FOV but the vids and clips posted in the ads were clearly far narrower than that. People buy the fake numbers not thinking about it just because the numbers sound good. 170 degrees FOV horizontally is equal to human vision and nobody has ever made a single-lens camera which could match that undistorted. The edges of the frames of my regular reading glasses are about 110-120 degrees horizontal FOV, yet nobody who wears glasses thinks they're excessively narrow in FOV. The aspect ratio also matters. The ultra-wide 21:9 being implemented gives a different diagonal FOV number than the previous taller one, yet there's been no change in lens or sensor and no width in horizontal FOV was added even though it now appears wider.

In my own testing I'm one of a few (and maybe the only one here) who actually measures horizontal FOV with instruments. The way my cams mount above my dashboard allows me to place a protractor exactly below the cam lens vertically where I can then check the numbers. Even done with great care I can easily be a couple degrees off. I've considered building a calibrated test jig for this measurement which would need to be rather long so the cam could be in focus where the markings are. If I tested more cams I might even do that, but the numbers I can find aren't of much use except to compare cams I've tested which are only a few of the many. Without testing every popular cam my numbers don't do much to help anyone.

Everyone here has always said to view the vids and not the specs if you want to know what a cam records for you because that's what you're buying a cam for- specs mean nothing. There are 1080p cams with clearer images than some 2K cams, and similar for some 2K cams versus some 4K cams. Some of the most expensive cams have poorer IQ than some budget cams do even though the specs are better. In the dashcam game you don't believe specs unless you know nothing about dashcams. Someone coming here asking the whole industry to change to more realistic specs clearly doesn't understand dashcams no matter what they claim to know.

The world doesn't change just because you want it to and only fools and visionaries even bother to try. Neither get results and visionaries are mighty rare.

Phil

 

[ursamajor]

I'm not accepting your reality as you are assuming what the numbers quoted are specifying when they don't actual make any mention of what the figure is meant to indicate, and you don't actually know, you have just assumed what it means, you're putting words in their mouths, I'm not suggesting that that's how things should be where people are left to assume or guess or whatever, it's confusing and I get that

I know what you say and this is full of... nonsens.
My reality is the physics reality. The FOV advertised should be the real FOV. As customer, I don't care what sensor+lens they used, but I do care about the real FOV and what they advertise IS NOT THE REAL FOV. And, to speak same language, the advertised FOV should be the HORIZONTAL one, not diagonal. We don't see in diagonals...
And, as possible customer, when I want to document me about a camera for buying and I read specs like in the picture below, should I guess what they wanted to mean with 140 degrees wide viewing angle? What you would understand reading such of specification?

Making a parallel with photo... if I buy a lens with 50 mm focal I know it has 50 mm focal (on full frame sensor@35 mm film) and I also know that on a crop sensor (for example Canon, or Nikon, or Olympus) the crop factor will multiply the focal and I will have an equivalent focal of 50xcrop factor (lower FOV). But in the case of dashcam I'm not forced to know (and it's completely useless for me) the focal@field of the lens and the crop factor of the sensor. Maybe would be better if they would advertise these things. But, instead, they advertise absolutely adopted/nonsens values (in romanian we have the expression "they took them out from the belly").

The world doesn't change just because you want it to and only fools and visionaries even bother to try. Neither get results and visionaries are mighty rare.

Phil

Wow, you're so "spiritual", but you're not smart at all when you're trying to offend me. Because I only feel offended by people smarter than me.
You forgot a category, dreamers. I am a dreamer and sometimes dreams come true.

PS: The discussion was civilized until you came... maybe would be better to keep it so.

 

[gse]

We are interested in the horizontal angle
119V3 ≈ 95°
129 Plus ≈ 112°

 

[ursamajor]

Good. Bravo!
I measured my A129+ with a protractor and I got about 120-125 degrees FOV. Maybe it could be a little error. Then I also calculated it few days ago, by reporting it to your measurements on V3 (as you stated, 95 degrees) using the rule of three, by comparing two pics. One of them from a video taken with my A129+, the other with a V3 of a user from a romanian forum, hometown fellow, in exactly same place from town.
1600pixels/1248pixels*95= about 121 degrees. 10 degrees of error/difference are perfectly understandable... and within the limits of the method.

 

[ursamajor]

Is there a place from where I could download previous firmwares for A129 Plus? I am in emailed conversation with a guy from Viofo, but he seems to not understand what and why I want.
Or maybe somebody has all the archive.
Thanks in advance!

 

[Nigel]

PS: one more thing... the real FOV is so easy to be measured. They have a bunch beta testers, they could ask them to measure the real angles and use the measurement in advertising the product. But they still prefer to advertise lies. What we are talking about?

The problem with that is if you ask 10 testers to measure it, you will get 10 different results!

By far the best way to deal with FOV is to put two cameras side by side in a car and compare the videos side by side.
In fact that is the only way to be accurate due to dashcam lenses having a lot of fisheye distortion, which distorts at different rates between lenses, so on some lenses the last few degrees are very compressed compared to others and may not be of much use for reading plates but very useful for extra FOV to show what is happening.

but he seems to not understand what and why I want.

Why do you want?
I don't think I would recommend using old firmware for what is quite a new model.

 

[ursamajor]

Why do you want?
I don't think I would recommend using old firmware for what is quite a new model.

I only want to test previous firmwares, because I have the suspicion that they modified the AE metering in last firmwares. I noticed increased luminosity at night and, as result, more noise, probably due to increased gain. Also, WDR works very, very bad, even with negative exposure compensations.

 

[Nigel]

I only want to test previous firmware, because I have the suspicion that they modified the metering in last firmwares. I noticed increased luminosity at night and, as result, more noise, probably due to increased gain. Also, WDR works very, very bad, even with negative exposure compensations.

Ignore the WDR setting, on some cameras it is better, but on Viofo cameras the standard setting is normally optimised to be as good as possible, the WDR setting is just an alternative that people can try.

You may be correct about a bit more noise, but it is always a compromise between various factors, and your view is likely to be a little different to that of others, so unless you want to use an old firmware permanently then it is not going to help.
The sensor used will always show some noise if it is optimised for reading plates at night, it becomes a compromise between noise and motion blur, with no correct solution.

Need a better sensor if you want a noise free night time camera with good plate readability.

 

[ursamajor]

No, IMX335 is a good enough sensor for me. I'm not absurd to pretend a noiseless sensor. To understand what I mean, look at the videos below. It's true, they were taken with A119V3 with an older firmware, but the same sensor. That old firmware used a default setting such that the image at night was much darker than in the following firmwares (see last video). WDR also worked flawlesly, as you can see in videos.
My suspicion is that they made those "optimisations" you mentioned above and destroyed the good balance between default exposure and WDR.

 

[Nigel]

That old firmware used a default setting such that the image at night was much darker than in the following firmwares (see last video).

But which is the best in that last video?

They are both perfectly valid options, some people will like the darker one, others will like the brighter, some will worry about noise, other people won't notice the noise.

I tend to choose a darker option so that there is less motion blur, but I do want to see that bicycle hidden in the dark shadows, so not too dark - there is no correct answer.

 

[ursamajor]

I like both, but I like the way WDR works there.
Compared to my A129+, its video w/o WDR is a little brighter, but when WDR is activated, the noise is much more than in A118V3 video.
I would prefer to have darker video when WDR is not active. I could increase myself EV value if I would really need it. But I would have also correctly working WDR.

 

[SawMaster]

PS: The discussion was civilized until you came... maybe would be better to keep it so.

Not at all. You started the mess here all by yourself, first by bashing Viofo over specs where his are a whole lot closer to realuity than many other manufacturers. And then you continued by bashing Jokiin whose cams specs are as close to reality as he can make them. He does his cams his way and honestly too, because that's what he believes is right- even if that costs him a sale.

You come here argumentative and hostile towards anyone who disagrees with you, then you claim to be civil. You come here claiming vast dashcam experience yet you don't seem to have ever noticed how cams advertised features and specs are almost never for real. You are a hypocrite and a cad who thinks you're always right.

I've held my tongue previously about how you treat other people in hopes that you might learn to take a more mellow approach, but you're too far gone to learn anything. So with this message, I'm done with you.

Phil

 

[kamkar]

I too prefer to compare 2 cameras side by side in car, i care not much for the exact FOV of the lens, but it seem like i prefer narrower lenses over wider ones.
Of course it can also be too narrow just i have never seen that in a factory dashcam maybe aside for the mobius A lens, same i have also not seen a too wide FOV lens but i have seen plenty a bit wider than what i would personally like.
But i am so used to what i like are often unique or none existing and so i will have to modify things to match what i like.

But dashcam makers should mention true numbers, and weather it is diagonal measurement or horizontal which i feel are the correct way.