Hit and Run--dash cam couldn't make out the license plate

[SawMaster]

It does when you want a clean clear video of an incident when its completely blurry or pixellated that you can't make a thing out....or at least the little details of it.....that you need or want for evidence....

We're not talking about cheap crappy cams like that here- at least I'm not. And TBH even my absolutely horrible G1W clone ($26 delivered!) did well enough to prove innocence even if it could barely get the plate number of a car stopped right in front of me in broad daylight Car placement and movement, lane markings, signage, traffic lights, taillights, brake lights, and turn signals were all displayed without any room for question. That's all you need to prove your innocence. Nice to have better though

You seem to just want to argue that all 1080p dashcams have crappy video when they don't Have a look at what Street Guardian does with their 1080p cams, better than some 2K and 4K cams get you Maybe you've only had a defective or junky cam to base your experiences on, I don't know, but I do know that the vast majority of dashcams in use right now are likely 1080p30, and most of their owners seem to be fairly happy with that level of performance. Given a good build with good FW and a decent bitrate 1080p30 can do quite well

As to FOV the difference between a 'box' measured diagonally versus horizontally should be easy to see or imagine; the diagonal will always be wider. While I think horizontal measurement is more meaningful for us, the industry standard for cameras is to measure diagonally so that;'s what we get. I've used a protractor to test many cams horizontal FOV and have found that even when allowing for the diagonal to be more, very few meet or even come close to their advertised value. It's well-known among experienced dashcammers to not believe those figures but instead to look at the videos and still-frames posted to see what you really get. That Garmin is certainly not close to 180 no matter what they claim. In fact few cams go past about 135 degrees diagonally in the vids you see from them with many popular ones doing more like 110- 120 degrees Still pretty good for our purposes no matter the claimed values being so inaccurate.

Phil

 

[jokiin]

Hmm yes, definitely 110 degrees tops.

check the video, you won't see 180°viewing angle, regardless of what some seller claims

 

[DD20]

A quick look on YT, and it is clearly not 180, it is fairly wide for a dashcam, maybe a bit too wide, but it is not 180.

With 180 you would clearly be able to see down the road we are joining at this T junction in both directions:
View attachment 55726

Are you sure that's a Garmin 66W? Looks like the video was stolen from this reviewer of this Innosinpo 170 degree fov dashcam.... https://images-na.ssl-images-amazon.com/images/I/A1KEN8cg1uS.mp4 ...or maybe it was the other way around since they blurred out the bottom left bit in the video which I presume would be Garmin's watermark, timestamp, GPS coordinates and speed....but that wouldn't make sense from a verified purchase....

I doubt it except for possibly some 'consumer protection' group or possibly an independent testing organization. Try searching for reviews and see how detailed they might be.

I found this guy who states:

What’s new with the Dash Cam 66W is its wide-angle lens, which shoots at 180 degrees, compared to 122 degrees with the Dash Cam 55. This is a significant difference and means your video includes a much wider view across the front of you vehicle. We found it was wide enough to see a little to the side of our car, which should help to capture footage of someone t-boning us at a junction, or a vehicle causing a collision by drifting into our lane.

Only notable detail would be being able to see the sides of the car....maybe he got a genuine 180 degree fov and all ele was like 110 degree fov max?

You seem to just want to argue that all 1080p dashcams have crappy video when they don't

Well I've seen a lot of youtube dash cam footages and they they look a little....crap....in my opinion of course which is where this 1080p cameras are crappy kind of vibe is coming from....

Have a look at what Street Guardian does with their 1080p cams, better than some 2K and 4K cams get you

Hmmm, I'm guessing this is the street guardian website?

As to FOV the difference between a 'box' measured diagonally versus horizontally should be easy to see or imagine; the diagonal will always be wider. While I think horizontal measurement is more meaningful for us, the industry standard for cameras is to measure diagonally so that;'s what we get.

What about vertical? The area between top and bottom? Doesn't this FOV area get mentioned? Or would that be the same as diagonal and horizontal or half of it?

I've used a protractor to test many cams horizontal FOV and have found that even when allowing for the diagonal to be more, very few meet or even come close to their advertised value.

It'd have to be a pretty big protractor.....so it's seen on camera and not the little handheld one you use to draw semi circles and straight lines with at school....

In fact few cams go past about 135 degrees diagonally in the vids you see from them with many popular ones doing more like 110- 120 degrees

So which ones go over 135 degrees?

Still pretty good for our purposes no matter the claimed values being so inaccurate.

So if the claim of 180 degree is false....wouldn't someone have had got them for false advertising?

check the video, you won't see 180°viewing angle, regardless of what some seller claims

Yes...but without a reference 180 degree fov as comparison, it's hard to see if it's 180 or 179 or some other value....

 

[DT MI]

I found this guy who states:

What’s new with the Dash Cam 66W is its wide-angle lens, which shoots at 180 degrees, compared to 122 degrees with the Dash Cam 55. This is a significant difference and means your video includes a much wider view across the front of you vehicle. We found it was wide enough to see a little to the side of our car, which should help to capture footage of someone t-boning us at a junction, or a vehicle causing a collision by drifting into our lane.

I don't see anywhere where he actually tested or measured it - just regurgitating the marketing claims. You're free to believe what you want but the video clip in post #80 of this thread clearly shows less than 180.

 

[DD20]

I don't see anywhere where he actually tested or measured it - just regurgitating the marketing claims

OH....well it sounded like he tested it....I guess not... Well in that case I have not found anyone who's tested or measured the 180 degree fov claim....my search results come empty on that regard unless I'm not searching right.... I mean there's this video that compares 140 to 180 if they really *are* 140 and 180 respectively but not using a giant 180 degree protractor like SawMaster would have used to test and measure accuracy of the supposed fov claim.... Hmmmm, I'll look some more tomorrow maybe I'm just too sleep at the moment to get the right content I need....

You're free to believe what you want but the video clip in post #80 of this thread clearly shows less than 180.

Well if there was a reference protractor of some sort overlayed onto the video that would help visualise the capture fov compared to its advertised fov....

 

[Nigel]

Are you sure that's a Garmin 66W? Looks like the video was stolen...

Yes, comes from:
https://kit.honestjohn.co.uk/reviews/review-garmin-66w/
, your link is the stolen one!

You don't want 180 degrees anyway unless you have enough resolution for it, which in my view is 4K minimum, otherwise you will have difficulty reading plates.

 

[DT MI]

...Well if there was a reference protractor of some sort overlayed onto the video that would help visualise the capture fov compared to its advertised fov....

Effectively there is - it's the 90 degree T-junction (interchange) in the image. From that it's obvious the horizontal FOV is not 180 degrees.

 

[SawMaster]

You don't need a huge protractor. All you need is to have a stillframe image on hand with enough reference points so that you can sight across a properly-oriented protractor to align with the edges of that still-frame shot. The tough part is protractor alignment, lots involved in that but it's all covered in 6th grade geometry classes. Being it's a comparative decision is why I claim + or - 2 degrees; I could be that far off but I seriously doubt any worse. In my job I frequently have to discover and measure angles to make things fit so I understand the process and am experienced with doing it.

It gives only a horizontal value the way I measure but that's what matters most. Every dashcam has plenty of vertical FOV even for overhead traffic lights. Getting a diagonal measure would much, much harder to do accurately with my simple gear and methods so I don't try, therefore I can't easily verify a claimed number, but I can tell when one is so far out that it's bogus. Aspect ratio plays a role here too, so two different diagonal numbers can actually have the same horizontal FOV angle A vertical measure which I could do coupled with some math could get me the diagonal number, but if I begin doing that or using my electronic angle finder (which BTW is accurate to +/- 0.05 degrees) then I'll have gone stark-raving mad and should be locked away in a rubber room Luckily my old van lets me get the measurement easily and fairly accurately with the center ridge in the hood, a measured center of the windshield at the top where I have a bracket mounted, and a mild windshield slope which puts part of the dashboard directly below the face of the cam lens so I can place my protractor accurately as I compare the image on my laptop to what I see across the protractor markings. Far from this easy in most vehicles.

My numbers suit our purpose by showing how far a cam 'sees' to either side from straight ahead, and darn few dashcams are installed perfectly centered and perfectly straight. Your car might be to one side of a lane or the other. It may not be straight in your lane. There are lots of tiny details which can change what a cam sees so + or - 2 degrees horizontal is plenty close enough. I've never heard of anyone using my numbers to check the horizontal FOV in their car before buying but they could if they wanted to. Most folks will use my numbers to make an educated guess or to compare cams. They won't be as far out as that Garmin claim I assure you.

The inflated numbers in advertising are just to sell cams to buyers who neither understand nor know how to verify FOV angles. The numbers I speak of are reasonably accurate as those with lots of dashcam experience know and understand. The dashcam game is full of false and misleading claims, believe them at your own risk but don't cry when you've been told the truth and won't believe it then find out too late that you were wrong.

Phil

 

[DD20]

Effectively there is - it's the 90 degree T-junction (interchange) in the image. From that it's obvious the horizontal FOV is not 180 degrees.

I suppose that's assuming the roads on the two ends are exactly opposite each other straight for a good few hundred meters and not curved around backwards or forwards....and of course camera placement would have to be dead center on the windscreen and not off at an angle or off to the right/left side but angled to point centered-ish. Also I believe aspect ratio would play a good part in lessoning a fish eye(Or this curved look on the edges) effect on a straight 180 degree fov. Assuming the road isn't curved, then yes I guess it doesn't look like 180 degree horizontally....hmmm.....someone should sue garmin for false advertising!

 

[Deleted member 73221]

Amazing. We created eye surgery but can't figure out how to make a lens that can capture a moving license plate. I'm sure if such a lens would cure corona, they'd figure it out in weeks

 

[kamkar]

It is not the lens, it is the sensors, they are simply not light sensitive enough to do it, and if they where we would probably need other lenses with a active aperture as otherwise too much light would get in and hit the sensor during the daytime.

Even the human eye are pretty limited, most of what we see is our brain elaborating on a blur as the human vision are only crisp / sharp in a very little field right in the middle of our field of view.

If you look into machine vision, which can detect and sort stuff at tremendous speeds, well where the camera look there will also be ample of light, otherwise the machine would see very little too.

 

[MarquisOfArgon]

How can you guys trust Chinese brands?

I am sorry, but either you don’t read the news or have blinds on.

Every Chinese app goes through servers that give full access to PLA hackers. No one has read US intelligence warnings on tiktok? like come on...

and worst part, those Chinese cameras are so short-lived. Before getting a GoPro, I tried one from Amazon- died after a week. Had perfect ratings. I gave it an honest review and a few days later the seller emailed me with a sob story and freebies to remove the review.

My GoPro also failed after a year- GoPro replaced it free of charge, no questions asked. I have a garmin fenix watch that is 7 years old and still going great, still gets updates, connection to GPS is flawless, still like new

you guys won’t know quality even if it stared you in the face

to the OP - trust me these guys are naive. Get the Garmin 56 or 66 from Amazon, try it at 1440 60 FPS. If you don’t like it, just return it. I bet you 2k 60fps would have captured your hit and run plate

another feature of garmin cameras that I absolutely love is the ability to give voice commands. I have both ok garmin and Alexa on my sat nav and ok garmin is way more responsive. You don’t have to fiddle with any buttons - just say “ok garmin save the video”

and never trust a Chinese company even if your life depended on it

Micky must live in quite some house if he totally avoids anything Chinese.

Just what is so special about your cam footage that you're so reluctant to let the Chinese see?

Maybe you should read a little less Daily Mail, drop the chip on the shoulder, and get out there into the big wide world and expand your horizons.

Oh, and by the way, my Nextbase 622 says designed in UK, made in China. And it's not very good. So who do we blame for that?

Xièxiè nǐ

 

[Nigel]

Oh, and by the way, my Nextbase 622 says designed in UK, made in China. And it's not very good. So who do we blame for that?

The company that puts (or allows use of) their brand name on the camera is the company you should hold accountable for good/bad design and quality. There are good and bad in all countries.

 

[kamkar]

Yeah.
It is funny people do not get it, you can get excellent quality from China too, but you dont get that for 9.99
And that is and alwasy have been the truth in any country, so no problem finding someone that take a 30 dollar product in china, slab a fancy name on it and then sell it for 100 dollars, you dont have to be Chinese to have that kind of business approach.

 

[Nigel]

Yeah.
It is funny people do not get it, you can get excellent quality from China too, but you dont get that for 9.99
And that is and alwasy have been the truth in any country, so no problem finding someone that take a 30 dollar product in china, slab a fancy name on it and then sell it for 100 dollars, you dont have to be Chinese to have that kind of business approach.

There is also the possibility of a brand being sold to a different country, who do you blame then?
For example Hasselblad is now owned by DJI of China; so do we blame China or Sweden for the excellent quality of their cameras?

 

[kamkar]

I just want one of the 7 hasselblad cameras tossed by Americans on the moon, that must be worth a nice penny

 

[Nigel]

I just want one of the 7 hasselblad cameras tossed by Americans on the moon, that must be worth a nice penny

A few more years and tourists will be visiting the moon on a treasure hunt for thoses valuables to help pay for their trip, and there will be a big fuss about not disturbing the archaeology but no moon police force to enforce it!

 

[MarquisOfArgon]

The company that puts (or allows use of) their brand name on the camera is the company you should hold accountable for good/bad design and quality. There are good and bad in all countries.

I was being facetious.

I work in oil and gas, our main fabrication partner is based in China and I'm a regularly visitor. I know better than most their strengths and weaknesses.

I hold Nextbase responsible for designing a duff camera, no one else.

My point was someone in this thread is borderline xenophobic.

 

[Nigel]

I work in oil and gas, our main fabrication partner is based in China and I'm a regularly visitor. I know better than most their strengths and weaknesses.

Yes I understood that, I was agreeing with you, although China is a very big place with many different peoples, as varied as Europe if not more.