Is the Mini 3’s photo quality as good as the Mavic 2?

[Wreckhunter]

And one more time, this is simply wrong; 48MP quad Bayer cameras have been used in phones for several years, and the Mini 3 Pro has one.

Quad Bayer sensors: what they are and what they are not

Marketing departments are trying to sell you on the wrong thing - resolution isn't why these sensors are good, it's the better image processing.

www.gsmarena.com

i do not and am not in the mood to engage with anyone on here.But check the stats on the mini 3 camera
and you will find it is in fact a true 12 mp sensor period,with 48mp trickery and a sales gimic.
it is not a 48 quad bayer camera as you stated above,it is a filter array.So in closing you are still in fact goofed
up here.
Please to not comment on this as I am done also.

 

[Lister]

One more time the mini 3 pro has a 12 MP sensor period,your getting your wires crossed and have it all backwards.
There is no camera drone currently that the public can get that has a camera with a true 48MP sensor.

Do you have access to a Mini 3? I suggest you shoot a JPG in the 48 MP mode, open the file in your favourite photo editor and inspect the pixel dimensions of the image. Then, report your findings here.

 

[RogerDH]

i do not and am not in the mood to engage with anyone on here.But check the stats on the mini 3 camera
and you will find it is in fact a true 12 mp sensor period,with 48mp trickery and a sales gimic.
it is not a 48 quad bayer camera as you stated above,it is a filter array.So in closing you are still in fact goofed
up here.
Please to not comment on this as I am done also.

Well, if you're not in the mood to engage with anyone on here, then I think you're the one who needs to stop commenting -- at least until you read the article I linked to, or Google your own -- there's no shortage of information about these sensors.

 

[Cymru]

Do you have access to a Mini 3? I suggest you shoot a JPG in the 48 MP mode, open the file in your favourite photo editor and inspect the pixel dimensions of the image. Then, report your findings here.

Thats not how it works. You have a 48mp image with the detail resolution of 12mp (and huge colour artefacts as a bonus)

 

[Lister]

Do you understand what's meant by "pixel dimensions" ?

 

[RogerDH]

Thats not how it works. You have a 48mp image with the detail resolution of 12mp (and huge colour artefacts as a bonus)

48 million photodiodes give you 48 million pixels of brightness data, which was the issue raised here. There are several reasons why these sensors can't be compared to larger 48MP sensors with conventional filtering, but that's a different matter. As I said, there is plenty of information available about what the sensors can and can't do.

 

[Lister]

48 million photodiodes give you 48 million pixels of brightness data,

Which is why I moaned about our inability to shoot true 48 MP monochrome images.

 

[Wreckhunter]

Do you have access to a Mini 3? I suggest you shoot a JPG in the 48 MP mode, open the file in your favourite photo editor and inspect the pixel dimensions of the image. Then, report your findings here.

Actually I have an air 2 with the 12mp sensor and 48mp quad bayer filter.The findings are and yes I have done the comparisons.
The 48mp setting,the photos are not close in quality that's why I never use it.
A question for you the fly app has a 48mp setting,if in fact the sensor was a true 48mp,then why does it have a 48 mode to select.
What would the regular sensor count be ,well lets see could it be 12 mp as is the true count.
Its not a mini 3 pro but it is the same setup.

 

[RogerDH]

Actually I have an air 2 with the 12mp sensor and 48mp quad bayer filter.The findings are and yes I have done the comparisons.
The 48mp setting,the photos are not close in quality that's why I never use it.
A question for you the fly app has a 48mp setting,if in fact the sensor was a true 48mp,then why does it have a 48 mode to select.
What would the regular sensor count be ,well lets see could it be 12 mp as is the true count.
Its not a mini 3 pro but it is the same setup.

Your mention of the Air 2 had me Googling about that camera, and I think I might have coincidentally come across at least one of the sources of this 12MP/48MP misinformation on DroneDJ.com:

DJI's marketing for Mavic Air 2 exaggerates specs that may be misleading

Does DJI overexaggerate specs about their drones? Are the 48MP photos on Mavic Air 2 truly 48MP or are they really shot on a 12MP sensor?

dronedj.com

Here's a quote from that article:

The Mavic Air 2 is able to capture the higher megapixels through the Quad Bayer Array. These sensors are also fairly popular among some of the smartphones that exist today. To achieve said quality (48MP), the true 12MP sensor breaks each pixel into 4 pixels. Here is a reference to an article on how this technique works. After all, a 48MP photo should be 2-4 times the size and quality of a 12MP image.

The "reference to an article" just happens to be the same one I gave you, and apparently this paragraph was missed:

A Quad Bayer filter is a bit of a misnomer as it’s actually the same as a regular Bayer filter. What really changes is not the filter but the sensor behind it – these new sensors put four pixels behind each color square instead of just one.

It's not 4 filters over 1 photodiode (which doesn't make sense, anyway), but 4 photodiodes under each filter.

 

[bradm75]

not even sure how they are doing it

f1.7. That wide of an aperture is going to let in a lot of light.

 

[KS-6]

Revisiting this thread after various firmware changes and DxO.

My original view still stands.
The Mavic 2 Pro wins hands down for photo quality - everything from range to detail in the image.

The mini 3 at 12mp is OK, especially after feeding through DxO but isnt better. The 48mp has far too many colour artefacts introduced to produce usable images at full scale.
There are also a LOT of lens distortions visible even after corrections and on 12mp (verticals on things like towers are obvious for example). Horizontal lines all curve towards the corners.

For video though the higher bit rate yields far smoother, less jittery video than my M2P can produce.

Interesting, because through my processing regimen I am getting very little in the way of color or noise gremlins and all this output is going to clients for magazine spreads, print adverting and other full resolution requirements.

I find it to be on par with my Mavic 2 Pro with that being said. I still prefer my EVO II Enterprise and my Mavic 3 Cine to it of course but for a bird that takes up the same room as an average medium speed prime in my bag, it is really quite fantastic.

 

[Cymru]

The colour shifts and artefacts make the images unacceptable for any sort of commercial use, even small online scenarios.

12mp is fine but absolutely no client or agency i have is going to accept images with those huge flaws.
The quad bayer introduces huge colour shifts on all bright, high contrast edges with detail. All the M3s do the same.
And as its not standard CA its a nightmare or impossible to remove.
Significantly better detail taking the 12mp and if needed, upsizing than the "48mp" and downsizing.

 

[Phantomrain.org]

Pixel peeping applies only if there is no wind in the sky that can alter the picture which is very rare , so what really becomes much more important is what drone can negotiate the wind better to take the most clean picture and or Panos and its not the Mini 3 Pro .

This is why the bigger drones like the Mavic 3 have the advantage as well as a bigger sensor , thus the Mavic 2 Pro is such an essential drone for many of the Pro's as it can negotiate the wind as can the Air 2S but the Mini 3 Pro and Mini 2 not so much.

So if you have to fly and shoot for 100 days in a row, its not the Mini 3 Pro you want in the bag.

Phantomrain.org
Gear to fly in the Rain. Land on the Water and shoot 100 days in a Row.

 

[Cymru]

Direct post of a 100% crop of an image showing artefacts hundreds of pixels long is not pixel peeping.
It has absolutely nothing to do with wind in the slightest. Its everything to do with the quad bayer processing.

You just used this thread as a thinly veiled attempt to sell snake oil. Again.

 

[Phantomrain.org]

My post was really directed to the OP , really was not denying your results of the Quad but those of us that enjoy the Panos , have a different take . .

Phantomrain.org
Gear to fly your drone in the RAIN. Land on Water ,

 

[Meta4]

Pixel peeping applies only if there is no wind in the sky that can alter the picture which is very rare , so what really becomes much more important is what drone can negotiate the wind better to take the most clean picture and or Panos and its not the Mini 3 Pro .

This is why the bigger drones like the Mavic 3 have the advantage as well as a bigger sensor , thus the Mavic 2 Pro is such an essential drone for many of the Pro's as it can negotiate the wind as can the Air 2S but the Mini 3 Pro and Mini 2 not so much.

Wind affects image quality ??
Now I've heard everything.

 

[KS-6]

View attachment 156969View attachment 156970

View attachment 156971


The colour shifts and artefacts make the images unacceptable for any sort of commercial use, even small online scenarios.

12mp is fine but absolutely no client or agency i have is going to accept images with those huge flaws.
The quad bayer introduces huge colour shifts on all bright, high contrast edges with detail. All the M3s do the same.
And as its not standard CA its a nightmare or impossible to remove.
Significantly better detail taking the 12mp and if needed, upsizing than the "48mp" and downsizing.

Interesting, I am not getting those magenta artifacts.

 

[KS-6]

Wind affects image quality ??
Now I've heard everything.

Actually, I have wondered about this as I do see image quality inconsistencies the most across all my drones when there is wind involved.

I can’t say that is the cause of it because that would take work that I am not really up to pouring time into but it is oddly coincidental.

 

[RogerDH]

Which is why I moaned about our inability to shoot true 48 MP monochrome images.

I'm not sure about this, but wouldn't you get that if you extract just the luminance from the DNG files?

 

[Cymru]

FWIW the JPG files produced on the "48" images dont have the artefacts, or at least not as obvious so it seems they're doing some processing to remove it in the JPG pathway which isnt being addressed on the DNG.