G90-7S (A7LA70 chipset + GPS) vs. Mini 0803 (A7LA50 chipset + GPS)

....the 1080 image is not as sharp as I would expect either and the night time image shows that there is dirt either on the lens or the glass.

....my Mini 0803 images tend to be a lot sharper than the page one 1296 image...
Lens & glass were clean. It was raining, as you'll see in the video.;)

I've tried to record footage in differing driving environments & speeds. For the 1296 daytime footage I was travelling at approx. 60 mph (100 kph), so both the G90-7S & Mini 0803 were having a realistic challenge.

What Mini 0803 FW were you using?
 
Last edited:
Lens & glass were clean. It was raining, as you'll see in the video.;)

I've tried to record footage in differing driving environments & speeds. For the 1296 footage I was travelling at approx. 60 mph (100 kph), so both the G90-7S & Mini 0803 we're having a realistic challenge.

What Mini 0803 FW were you using?
I was using 08/12 for images 2,3,4 and 06/11 for the rest, I think.

I understand it was a realistic challenge, but I don't think that night time starburst on the street lamp was caused by rain. The main difference between your 1296 image and mine is the reflections off the glass, you can't actually see through the reflection of the dash vent on yours for the 0801 but you can for the G90 which I think makes it an unfair comparison, although I don't see why one would have more reflection problem than the other, can the camera be blamed for the reflections?
 
I was using 08/12 for images 2,3,4 and 06/11 for the rest, I think.

....you can't actually see through the reflection of the dash vent on yours for the 0801 but you can for the G90 which I think makes it an unfair comparison, although I don't see why one would have more reflection problem than the other, can the camera be blamed for the reflections?
Thanks for the info on the Mini 0803 FW you'd used :)

Your Mini 0803 appears to have a sharper focus on the RHS compared to the LHS.

I don't understand why you think it has been an unfair comparison.:confused: There's 1 minute video footage from each dash cam, taken on the same journeys at the same time.:)
As expected slower speed footage gives a better image in both, as the screen shots and videos show.

It's impossible to have the G90-7S & Mini 0803 positioned/angled identically, so there's bound to be a slight difference in reflections. They were positioned/angled as closely as I could get them.

That particular screenshot was chosen because of the details being picked up, by both dash cams, in a night time/streetlight condition.
It was certainly wasn't chosen because it showed the Mini 0803 'starburst'.;)
 
Last edited:
Your Mini 0803 appears to have a sharper focus on the RHS compared to the LHS.
Yes, I suspect that may be due to the curvature of the windscreen glass, I've noticed that Mini 0803s in LHD cars which get located the opposite side of the mirror tend to be the other way around and I've not noticed it when used outside the car...

I don't understand why you think it has been an unfair comparison.:confused:
...
It was certainly wasn't chosen because it showed the Mini 0803 'starburst'.;)
But do your images accurately represent what people should expect from the two cameras?

My Mini 0803 doesn't show any "moonburst" in my night time images so if your glass really is clean then why does yours show "lightburst" while your G90 doesn't?

I don't know the answer so I'm not sure if your test is fair or not, I do know that my experience of Mini 0803s is different to what is shown in your images.
 
....My Mini 0803 doesn't show any "moonburst" in my night time images so if your glass really is clean then why does yours show "lightburst" while your G90 doesn't?

I don't know the answer so I'm not sure if your test is fair or not, I do know that my experience of Mini 0803s is different to what is shown in your images.
The video footage is 'as recorded', i.e., the images have not been altered in any way. The screen shots are taken from the minute long video footage, some have been slightly cropped as mentioned earlier. The inside of the windscreen and both lenses were cleaned in the same way at the same time. The outside of the windscreen may have some dirt or insect splatter, but that's real life, it was the same conditions for both dash cams. I have made every effort to make sure all the comparisons were fair and representative of typical 'normal' usage.

Your Mini 0803 night time shots (in Post No.79) don't appear to be from typical dash cam usage; there are no street lights, they appear to have been taken whilst stationary and are of slow moving or static objects. Are they even taken whilst in your car?? :confused:

The 1296 night time screenshot was chosen because it shows how the two dash cams pick up images in a difficult lighting situation, e.g., the flowers in the pot & wall details. There is little difference between the two dash cams in these areas.

Have you watched the various night time videos through to see whether the "moonburst"/"lightburst"/"starburst" happens elsewhere on either the G90-7S or Mini 0803?o_O To save you time looking, it happens quite a lot in the Mini 0803 footage, I hadn't noticed that until you kindly highlighted it....:rolleyes: As I've said above, it doesn't appear to affect footage recorded in the main 'area of interest' (in the unfortunate event of an incident), i.e., at road level.:)

....But do your images accurately represent what people should expect from the two cameras?
Yes, I believe they do. If you are not sure about the screen shots then watch the videos in Sections 1-5, that's precisely why I'd posted links to them.;)

I've tried to provide an unbiased, balanced & fair comparison between the G90-7S & Mini 0803 and let others decide which they prefer.:D
 
Last edited:
0803 night tests ("poor" results) ...

Published on Aug 20, 2014 (night test @ 6:20)

Published on Jul 18, 2014 ( "lightburst" examples)

Published on Jul 10, 2014

 
0803 night tests ("poor" results) ...
Thanks for those links.

It would appear that my Mini 0803 screen shots & video footage are fairly representative. Actually the night time video footage from my Mini 0803 is better than that shown in those YouTube clips.;)
 
Last edited:
Here is an example why I am not satisfied with my mini 0803:

Frame 2014.09.04 17-35-08 (0-00-08.273).jpg

As you can see, I used best conditions for the test: the camera and the objects are still standing, it is a sunny day and there is no window between the camera and the objects.
Have a look at the car on the left, the distance is 9m / 29 ft only and I can hardly identify the number plate.
In an other threat someone said, to be able to do that the max. distance is about 50 ft when recording in full hd, but this is super hd and nearly half the distance. Very disappointing.

With my eyes I can even identify the numbers of the black car on the right without problems, the distance is about 25m / 82 ft. Ok, I have expected that this distance is too much for the cam, because of the wide angle... but the numbers on the nearest car I expected to be readable. The number plate in reality is so near and so big, that even a half-blind grandpa can read it.

And the camera is getting very hot, I don't know if this has an effect to its life expectancy.

Should I send it back to Amazon, what do you think? I prefer the size and mount of the mini over the G90...

Another question: This cam has autofocus? I don't believe that, although the setting "AF Meter Mode: Centre (default) or Spot" implements that. But if it has autofocus, you normally can see the action of the camera to focus to an object and that takes a second or so normally. What do you think, is the setting a fake? BTW, I tried both settings - center an spot - and couldn't see a difference.
 
Last edited:
With my eyes I can even identify the numbers of the black car on the right without problems,
The plate on that car is 6 pixels high, that is about 4 pixels for the text which is not enough even if you tried to fit them in by hand, for the car on the left there are about 7 pixels for the text which is right at the limit for representing text if you are doing it by hand - ie the mini 0803 setup menu! it's not done a bad job.

You need to compromise either with a significantly narrower field of view or more resolution if you want to read plates at that distance
 
Here is an example why I am not satisfied with my mini 0803:
...Have a look at the car on the left, the distance is 9m / 29 ft only and I can hardly identify the number plate.
In an other threat someone said, to be able to do that the max. distance is about 50 ft when recording in full hd, but this is super hd and nearly half the distance. Very disappointing.
I don't think that either my G90-7S or Mini 0803 would be able to pick up number plates at either of those distances. It is much nearer than that.

...even a half-blind grandpa can read it.
LOL - Never been called that before! :p

...This cam has autofocus?
No, it has a fixed lens/focus. Focal point is determined in the factory before it leaves.
 
Last edited:
Ok, then my expectations were too high :)
That's for your quick answers, guys!

you'd need a much narrower lens to get a longer focal distance, even then I can't see you getting 9 meters, don't know about you but if a car is 9 meters away from me at best I don't see it as being a threat, anything that hits your car is going to be closer than that. assuming you don't live in a warzone
 
I am living in a big metropole in the west of Germany called "Ruhrgebiet", an accumulation of several big cities close together, and yes - its a warzone according to the situation on the highways :p

Much too much traffic in combination with (mostly) no speed limits. On the right lane you have trucks driving 80 km/h, on the left lane you have Porsche drivers at 200 km/h. But the good thing is: It's not boring and you don't get in danger to fall asleep ;)
 
Last edited:
I've just watched the vids now I'm home and noticed the 2560x1080p footage is behaving differently than I'd expected - I'd assumed it would basically give you extra info at the sides, but from looking at the clips of the front of your car it's more like they're using the same sensor area as the 1920x1080 video and cropping the top and bottom:

g90-wide.jpg

I'm not sure if the cam was moved but it looks like the 2560 stuff is cropped and scaled 1920 footage like this:

g90-wide2.jpg

Rather than using the same sensor area height and going out to the sides, like this LG pic shows:

lg-29ea93-21-9-vs-16-9-544x344.jpg


I'd assumed the extra width would give you more at the sides but it doesn't actually look like that's the case.
 
It means that you can get better resolution while keeping the same data rate, at the expense of loosing the top and bottom, assuming that the native resolution of the sensor is higher than both formats which is likely.

Who wants a video of their bonnet anyway! Mind you, if I had more sky in mine instead of bonnet then I would have got a video of an aeroplane flying backwards last week, at least with a negative ground speed.
 
You've misunderstood me there mate - I'm meaning more if they were passing through the 1920x1080 centre of the sensor for video footage then the raw sensor area used for 2560x1080p footage should have extra width but it doesn't - so they're just rescaling the pic which isn't optimal.

It's something they can potentially improve via a software change.
 
the 2560x1080p footage is behaving differently than I'd expected.
are you speaking about 0803 or G90? I made a quick test using my G90 A7LA70 and the result is similar as yours. It looks like thieving.

It's something they can potentially improve via a software change.
Or maybe they can't because of the lenses. So instead of improving the hardware, they artificially created a new resolution by cropping the image at 1920x1080 and increasing the bitrate.
 
Back
Top